[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-amrzv--awesome-colab-notebooks":3,"tool-amrzv--awesome-colab-notebooks":61},[4,18,26,36,44,53],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":17},4358,"openclaw","openclaw\u002Fopenclaw","OpenClaw 是一款专为个人打造的本地化 AI 助手，旨在让你在自己的设备上拥有完全可控的智能伙伴。它打破了传统 AI 助手局限于特定网页或应用的束缚，能够直接接入你日常使用的各类通讯渠道，包括微信、WhatsApp、Telegram、Discord、iMessage 等数十种平台。无论你在哪个聊天软件中发送消息，OpenClaw 都能即时响应，甚至支持在 macOS、iOS 和 Android 设备上进行语音交互，并提供实时的画布渲染功能供你操控。\n\n这款工具主要解决了用户对数据隐私、响应速度以及“始终在线”体验的需求。通过将 AI 部署在本地，用户无需依赖云端服务即可享受快速、私密的智能辅助，真正实现了“你的数据，你做主”。其独特的技术亮点在于强大的网关架构，将控制平面与核心助手分离，确保跨平台通信的流畅性与扩展性。\n\nOpenClaw 非常适合希望构建个性化工作流的技术爱好者、开发者，以及注重隐私保护且不愿被单一生态绑定的普通用户。只要具备基础的终端操作能力（支持 macOS、Linux 及 Windows WSL2），即可通过简单的命令行引导完成部署。如果你渴望拥有一个懂你",349277,3,"2026-04-06T06:32:30",[13,14,15,16],"Agent","开发框架","图像","数据工具","ready",{"id":19,"name":20,"github_repo":21,"description_zh":22,"stars":23,"difficulty_score":10,"last_commit_at":24,"category_tags":25,"status":17},3808,"stable-diffusion-webui","AUTOMATIC1111\u002Fstable-diffusion-webui","stable-diffusion-webui 是一个基于 Gradio 构建的网页版操作界面，旨在让用户能够轻松地在本地运行和使用强大的 Stable Diffusion 图像生成模型。它解决了原始模型依赖命令行、操作门槛高且功能分散的痛点，将复杂的 AI 绘图流程整合进一个直观易用的图形化平台。\n\n无论是希望快速上手的普通创作者、需要精细控制画面细节的设计师，还是想要深入探索模型潜力的开发者与研究人员，都能从中获益。其核心亮点在于极高的功能丰富度：不仅支持文生图、图生图、局部重绘（Inpainting）和外绘（Outpainting）等基础模式，还独创了注意力机制调整、提示词矩阵、负向提示词以及“高清修复”等高级功能。此外，它内置了 GFPGAN 和 CodeFormer 等人脸修复工具，支持多种神经网络放大算法，并允许用户通过插件系统无限扩展能力。即使是显存有限的设备，stable-diffusion-webui 也提供了相应的优化选项，让高质量的 AI 艺术创作变得触手可及。",162132,"2026-04-05T11:01:52",[14,15,13],{"id":27,"name":28,"github_repo":29,"description_zh":30,"stars":31,"difficulty_score":32,"last_commit_at":33,"category_tags":34,"status":17},1381,"everything-claude-code","affaan-m\u002Feverything-claude-code","everything-claude-code 是一套专为 AI 编程助手（如 Claude Code、Codex、Cursor 等）打造的高性能优化系统。它不仅仅是一组配置文件，而是一个经过长期实战打磨的完整框架，旨在解决 AI 代理在实际开发中面临的效率低下、记忆丢失、安全隐患及缺乏持续学习能力等核心痛点。\n\n通过引入技能模块化、直觉增强、记忆持久化机制以及内置的安全扫描功能，everything-claude-code 能显著提升 AI 在复杂任务中的表现，帮助开发者构建更稳定、更智能的生产级 AI 代理。其独特的“研究优先”开发理念和针对 Token 消耗的优化策略，使得模型响应更快、成本更低，同时有效防御潜在的攻击向量。\n\n这套工具特别适合软件开发者、AI 研究人员以及希望深度定制 AI 工作流的技术团队使用。无论您是在构建大型代码库，还是需要 AI 协助进行安全审计与自动化测试，everything-claude-code 都能提供强大的底层支持。作为一个曾荣获 Anthropic 黑客大奖的开源项目，它融合了多语言支持与丰富的实战钩子（hooks），让 AI 真正成长为懂上",143909,2,"2026-04-07T11:33:18",[14,13,35],"语言模型",{"id":37,"name":38,"github_repo":39,"description_zh":40,"stars":41,"difficulty_score":32,"last_commit_at":42,"category_tags":43,"status":17},2271,"ComfyUI","Comfy-Org\u002FComfyUI","ComfyUI 是一款功能强大且高度模块化的视觉 AI 引擎，专为设计和执行复杂的 Stable Diffusion 图像生成流程而打造。它摒弃了传统的代码编写模式，采用直观的节点式流程图界面，让用户通过连接不同的功能模块即可构建个性化的生成管线。\n\n这一设计巧妙解决了高级 AI 绘图工作流配置复杂、灵活性不足的痛点。用户无需具备编程背景，也能自由组合模型、调整参数并实时预览效果，轻松实现从基础文生图到多步骤高清修复等各类复杂任务。ComfyUI 拥有极佳的兼容性，不仅支持 Windows、macOS 和 Linux 全平台，还广泛适配 NVIDIA、AMD、Intel 及苹果 Silicon 等多种硬件架构，并率先支持 SDXL、Flux、SD3 等前沿模型。\n\n无论是希望深入探索算法潜力的研究人员和开发者，还是追求极致创作自由度的设计师与资深 AI 绘画爱好者，ComfyUI 都能提供强大的支持。其独特的模块化架构允许社区不断扩展新功能，使其成为当前最灵活、生态最丰富的开源扩散模型工具之一，帮助用户将创意高效转化为现实。",107888,"2026-04-06T11:32:50",[14,15,13],{"id":45,"name":46,"github_repo":47,"description_zh":48,"stars":49,"difficulty_score":32,"last_commit_at":50,"category_tags":51,"status":17},4721,"markitdown","microsoft\u002Fmarkitdown","MarkItDown 是一款由微软 AutoGen 团队打造的轻量级 Python 工具，专为将各类文件高效转换为 Markdown 格式而设计。它支持 PDF、Word、Excel、PPT、图片（含 OCR）、音频（含语音转录）、HTML 乃至 YouTube 链接等多种格式的解析，能够精准提取文档中的标题、列表、表格和链接等关键结构信息。\n\n在人工智能应用日益普及的今天，大语言模型（LLM）虽擅长处理文本，却难以直接读取复杂的二进制办公文档。MarkItDown 恰好解决了这一痛点，它将非结构化或半结构化的文件转化为模型“原生理解”且 Token 效率极高的 Markdown 格式，成为连接本地文件与 AI 分析 pipeline 的理想桥梁。此外，它还提供了 MCP（模型上下文协议）服务器，可无缝集成到 Claude Desktop 等 LLM 应用中。\n\n这款工具特别适合开发者、数据科学家及 AI 研究人员使用，尤其是那些需要构建文档检索增强生成（RAG）系统、进行批量文本分析或希望让 AI 助手直接“阅读”本地文件的用户。虽然生成的内容也具备一定可读性，但其核心优势在于为机器",93400,"2026-04-06T19:52:38",[52,14],"插件",{"id":54,"name":55,"github_repo":56,"description_zh":57,"stars":58,"difficulty_score":10,"last_commit_at":59,"category_tags":60,"status":17},4487,"LLMs-from-scratch","rasbt\u002FLLMs-from-scratch","LLMs-from-scratch 是一个基于 PyTorch 的开源教育项目，旨在引导用户从零开始一步步构建一个类似 ChatGPT 的大型语言模型（LLM）。它不仅是同名技术著作的官方代码库，更提供了一套完整的实践方案，涵盖模型开发、预训练及微调的全过程。\n\n该项目主要解决了大模型领域“黑盒化”的学习痛点。许多开发者虽能调用现成模型，却难以深入理解其内部架构与训练机制。通过亲手编写每一行核心代码，用户能够透彻掌握 Transformer 架构、注意力机制等关键原理，从而真正理解大模型是如何“思考”的。此外，项目还包含了加载大型预训练权重进行微调的代码，帮助用户将理论知识延伸至实际应用。\n\nLLMs-from-scratch 特别适合希望深入底层原理的 AI 开发者、研究人员以及计算机专业的学生。对于不满足于仅使用 API，而是渴望探究模型构建细节的技术人员而言，这是极佳的学习资源。其独特的技术亮点在于“循序渐进”的教学设计：将复杂的系统工程拆解为清晰的步骤，配合详细的图表与示例，让构建一个虽小但功能完备的大模型变得触手可及。无论你是想夯实理论基础，还是为未来研发更大规模的模型做准备",90106,"2026-04-06T11:19:32",[35,15,13,14],{"id":62,"github_repo":63,"name":64,"description_en":65,"description_zh":66,"ai_summary_zh":66,"readme_en":67,"readme_zh":68,"quickstart_zh":69,"use_case_zh":70,"hero_image_url":71,"owner_login":72,"owner_name":73,"owner_avatar_url":74,"owner_bio":75,"owner_company":76,"owner_location":77,"owner_email":75,"owner_twitter":75,"owner_website":78,"owner_url":79,"languages":80,"stars":85,"forks":86,"last_commit_at":87,"license":88,"difficulty_score":89,"env_os":90,"env_gpu":91,"env_ram":92,"env_deps":93,"category_tags":107,"github_topics":108,"view_count":32,"oss_zip_url":75,"oss_zip_packed_at":75,"status":17,"created_at":124,"updated_at":125,"faqs":126,"releases":137},5092,"amrzv\u002Fawesome-colab-notebooks","awesome-colab-notebooks","Collection of google colaboratory notebooks for fast and easy experiments","awesome-colab-notebooks 是一个专为机器学习实验打造的 Google Colab 笔记本精选合集。它汇集了来自全球顶尖机构（如 Google、Meta、伯克利等）的高质量代码资源，涵盖从大模型智能体、交互式图像分割到强化学习课程作业等前沿领域。\n\n对于许多开发者和研究人员而言，配置复杂的深度学习环境往往耗时费力，且复现论文结果门槛较高。awesome-colab-notebooks 直击这一痛点，提供了“开箱即用”的实验环境。用户无需在本地安装繁琐的依赖库或担心显卡资源不足，只需点击链接即可在云端免费 GPU 上直接运行和修改代码，极大地缩短了从想法到验证的周期。\n\n这套资源特别适合 AI 研究者、数据科学家以及希望快速上手新技术的开发者。无论是需要复现最新论文算法的科研人员，还是想要通过实战案例学习大模型应用的学生，都能在这里找到对应的模板。其核心亮点在于不仅整理了热门开源项目（如 LangGraph、Qwen3-Omni 等）的演示笔记，还持续追踪社区趋势，确保内容的时效性与实用性，是探索人工智能领域的高效加速器。","[![Hits](https:\u002F\u002Fhits.seeyoufarm.com\u002Fapi\u002Fcount\u002Fincr\u002Fbadge.svg?url=https:\u002F\u002Fgithub.com\u002Famrzv\u002Fawesome-colab-notebooks)](https:\u002F\u002Fhits.seeyoufarm.com)\n![awesome-colab-notebooks](https:\u002F\u002Fcount.getloli.com\u002Fget\u002F@awesome-colab-notebooks?theme=rule34)\n\n[![Word Cloud](images\u002Fcloud.svg)](images\u002Fcloud.svg)\n\nThe page might not be rendered properly. Please open [README.md](https:\u002F\u002Fgithub.com\u002Famrzv\u002Fawesome-colab-notebooks\u002Fblob\u002Fmain\u002FREADME.md) file directly\n# Awesome colab notebooks collection for ML experiments\n## Trending\n| repositories | papers | packages |\n|---|---|---|\n| \u003Cul>\u003Cli>agent-starter-pack\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FGoogleCloudPlatform\u002Fagent-starter-pack?style=social)](https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fagent-starter-pack)\u003C\u002Fli> \u003Cli>PaddleHub\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPaddlePaddle\u002FPaddleHub?style=social)](https:\u002F\u002Fgithub.com\u002FPaddlePaddle\u002FPaddleHub)\u003C\u002Fli> \u003Cli>ritm-interactive-segmentation\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsupervisely-ecosystem\u002Fritm-interactive-segmentation?style=social)](https:\u002F\u002Fgithub.com\u002Fsupervisely-ecosystem\u002Fritm-interactive-segmentation)\u003C\u002Fli> \u003Cli>verl\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fvolcengine\u002Fverl?style=social)](https:\u002F\u002Fgithub.com\u002Fvolcengine\u002Fverl)\u003C\u002Fli> \u003Cli>Qwen3-Omni\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FQwenLM\u002FQwen3-Omni?style=social)](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen3-Omni)\u003C\u002Fli> \u003Cli>dinov3\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fdinov3?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdinov3)\u003C\u002Fli> \u003Cli>wmar\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fwmar?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fwmar)\u003C\u002Fli> \u003Cli>cwm\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fcwm?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fcwm)\u003C\u002Fli> \u003Cli>circle-guard-bench\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwhitecircle-ai\u002Fcircle-guard-bench?style=social)](https:\u002F\u002Fgithub.com\u002Fwhitecircle-ai\u002Fcircle-guard-bench)\u003C\u002Fli> \u003Cli>SAELens\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FjbloomAus\u002FSAELens?style=social)](https:\u002F\u002Fgithub.com\u002FjbloomAus\u002FSAELens)\u003C\u002Fli> \u003Cli>ARENA_3.0\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcallummcdougall\u002FARENA_3.0?style=social)](https:\u002F\u002Fgithub.com\u002Fcallummcdougall\u002FARENA_3.0)\u003C\u002Fli> \u003Cli>TabPFN\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fautoml\u002FTabPFN?style=social)](https:\u002F\u002Fgithub.com\u002Fautoml\u002FTabPFN)\u003C\u002Fli> \u003Cli>vjepa2\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fvjepa2?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fvjepa2)\u003C\u002Fli> \u003Cli>opik\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcomet-ml\u002Fopik?style=social)](https:\u002F\u002Fgithub.com\u002Fcomet-ml\u002Fopik)\u003C\u002Fli> \u003Cli>alphaevolve_results\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Falphaevolve_results?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Falphaevolve_results)\u003C\u002Fli> \u003Cli>SwanLab\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FSwanHubX\u002FSwanLab?style=social)](https:\u002F\u002Fgithub.com\u002FSwanHubX\u002FSwanLab)\u003C\u002Fli> \u003Cli>homework_fall2023\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fberkeleydeeprlcourse\u002Fhomework_fall2023?style=social)](https:\u002F\u002Fgithub.com\u002Fberkeleydeeprlcourse\u002Fhomework_fall2023)\u003C\u002Fli> \u003Cli>langgraph\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Flangchain-ai\u002Flanggraph?style=social)](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraph)\u003C\u002Fli> \u003Cli>sglang\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsgl-project\u002Fsglang?style=social)](https:\u002F\u002Fgithub.com\u002Fsgl-project\u002Fsglang)\u003C\u002Fli> \u003Cli>part_1_ml_cv\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FDeepLearningSchool\u002Fpart_1_ml_cv?style=social)](https:\u002F\u002Fgithub.com\u002FDeepLearningSchool\u002Fpart_1_ml_cv)\u003C\u002Fli> \u003Cli>felix\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsimonmalberg\u002Ffelix?style=social)](https:\u002F\u002Fgithub.com\u002Fsimonmalberg\u002Ffelix)\u003C\u002Fli> \u003Cli>prompt-eng-interactive-tutorial\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fanthropics\u002Fprompt-eng-interactive-tutorial?style=social)](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fprompt-eng-interactive-tutorial)\u003C\u002Fli> \u003Cli>presidio\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002Fpresidio?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpresidio)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>trlX\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0ece95190254.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002F2023.emnlp-main.530)\u003C\u002Fli> \u003Cli>LLaMA Factory\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0b29ad8610c4.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002F2024.acl-demos.38)\u003C\u002Fli> \u003Cli>AWQ\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_260e9c4ff296.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3714983.3714987)\u003C\u002Fli> \u003Cli>EAT\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2fe6fd6c5a7e.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.02069)\u003C\u002Fli> \u003Cli>Fine-tuning a BERT\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1d8bb9ef5451.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002FN19-1423)\u003C\u002Fli> \u003Cli>Classify text with BERT\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1d8bb9ef5451.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002FN19-1423)\u003C\u002Fli> \u003Cli>Panini-Net\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_dc79c78aa8d7.png)](https:\u002F\u002Fdoi.org\u002F10.1609\u002Faaai.v36i3.20159)\u003C\u002Fli> \u003Cli>Gaussian Splatting\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0c34c9f6bf85.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3592433)\u003C\u002Fli> \u003Cli>GraphCast\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_95a3b8c45abb.png)](https:\u002F\u002Fdoi.org\u002F10.1126\u002Fscience.adi2336)\u003C\u002Fli> \u003Cli>BiRefNet\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_b8af1a40a030.png)](https:\u002F\u002Fdoi.org\u002F10.26599\u002FAIR.2024.9150038)\u003C\u002Fli> \u003Cli>SAHI\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_09c209fb3be9.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICIP46576.2022.9897990)\u003C\u002Fli> \u003Cli>DifFace\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_28a1a4fb4867.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTPAMI.2024.3432651)\u003C\u002Fli> \u003Cli>py-irt\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9acfca034502.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002F2021.acl-long.346)\u003C\u002Fli> \u003Cli>FreeInit\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cba6c3f73678.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-72646-0_22)\u003C\u002Fli> \u003Cli>NeRViS\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fd196cdcdbc9.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00230)\u003C\u002Fli> \u003Cli>LIDA\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_14e441d2aa9f.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002F2023.acl-demo.11)\u003C\u002Fli> \u003Cli>AST\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_bb66827c48fd.png)](https:\u002F\u002Fdoi.org\u002F10.21437\u002FInterspeech.2021-698)\u003C\u002Fli> \u003Cli>Geometry-Free View Synthesis\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9f50cfdb8dc3.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.01409)\u003C\u002Fli> \u003Cli>AudioLM\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_30e549055623.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTASLP.2023.3288409)\u003C\u002Fli> \u003Cli>deep-significance\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fd26b66278d2.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002Fp19-1266)\u003C\u002Fli> \u003Cli>VRT\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_aedd82489829.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTIP.2024.3372454)\u003C\u002Fli> \u003Cli>GLIP\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_085ff51929b8.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01069)\u003C\u002Fli> \u003Cli>OWL-ViT\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9d15beeb274f.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20080-9_42)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>sam3\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fsam3?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fsam3\u002F)\u003C\u002Fli> \u003Cli>xminigrid\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fxminigrid?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fxminigrid\u002F)\u003C\u002Fli> \u003Cli>sglang\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fsglang?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fsglang\u002F)\u003C\u002Fli> \u003Cli>composer\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fcomposer?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fcomposer\u002F)\u003C\u002Fli> \u003Cli>a2a-sdk\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fa2a-sdk?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fa2a-sdk\u002F)\u003C\u002Fli> \u003Cli>dopamine-rl\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fdopamine-rl?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fdopamine-rl\u002F)\u003C\u002Fli> \u003Cli>google-cloud-discoveryengine\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fgoogle-cloud-discoveryengine?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fgoogle-cloud-discoveryengine\u002F)\u003C\u002Fli> \u003Cli>imagededup\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fimagededup?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fimagededup\u002F)\u003C\u002Fli> \u003Cli>xmanager\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fxmanager?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fxmanager\u002F)\u003C\u002Fli> \u003Cli>agent-starter-pack\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fagent-starter-pack?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fagent-starter-pack\u002F)\u003C\u002Fli> \u003Cli>torchgeo\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Ftorchgeo?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Ftorchgeo\u002F)\u003C\u002Fli> \u003Cli>rl-games\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Frl-games?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Frl-games\u002F)\u003C\u002Fli> \u003Cli>swanlab\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fswanlab?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fswanlab\u002F)\u003C\u002Fli> \u003Cli>transformer-lens\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Ftransformer-lens?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Ftransformer-lens\u002F)\u003C\u002Fli> \u003Cli>scikit-video\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fscikit-video?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fscikit-video\u002F)\u003C\u002Fli> \u003Cli>google-genai\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fgoogle-genai?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fgoogle-genai\u002F)\u003C\u002Fli> \u003Cli>tensorflow-graphics\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Ftensorflow-graphics?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Ftensorflow-graphics\u002F)\u003C\u002Fli> \u003Cli>neural-tangents\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fneural-tangents?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fneural-tangents\u002F)\u003C\u002Fli> \u003Cli>tensorflow-gnn\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Ftensorflow-gnn?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Ftensorflow-gnn\u002F)\u003C\u002Fli> \u003Cli>opik\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fopik?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fopik\u002F)\u003C\u002Fli> \u003Cli>sae-lens\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fsae-lens?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fsae-lens\u002F)\u003C\u002Fli> \u003Cli>csbdeep\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fcsbdeep?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fcsbdeep\u002F)\u003C\u002Fli> \u003Cli>langgraph\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Flanggraph?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Flanggraph\u002F)\u003C\u002Fli>\u003C\u002Ful> |\n## Courses\n\u003Cdetails>\n\u003Csummary>COURSES\u003C\u002Fsummary>\n\n| name | description | authors | links | colaboratory | update |\n|------|-------------|:--------|:------|:------------:|:------:|\n| LLM Engineering Essentials course | 12-week course, created by experts from academia and industry, is designed specifically for developers and engineers | \u003Cul>\u003Cli>[Stanislav Fedotov](https:\u002F\u002Fgithub.com\u002Fst-fedotov)\u003C\u002Fli> \u003Cli>[Alexey Bukhtiyarov](https:\u002F\u002Fgithub.com\u002Fleshanbog)\u003C\u002Fli> \u003Cli>[Nikita Pavlichenko](https:\u002F\u002Fwww.pavlichenko.info\u002F)\u003C\u002Fli> \u003Cli>[Sergei Petrov](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsergei-petrov-12589210b\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Sergei Skvortsov](https:\u002F\u002Fgithub.com\u002Fsouthfreebird)\u003C\u002Fli> \u003Cli>[Alex Umnov](https:\u002F\u002Fgithub.com\u002FAlexUmnov)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNebius-Academy\u002FLLM-Engineering-Essentials?style=social)](https:\u002F\u002Fgithub.com\u002FNebius-Academy\u002FLLM-Engineering-Essentials) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002Fkarpathy\u002Fstatus\u002F1886192184808149383)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Facademy.nebius.com\u002Fllm-engineering-essentials)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FNebius-Academy\u002FLLM-Engineering-Essentials\u002Fblob\u002Fmain\u002Ftopic1\u002F1.1_intro_to_llm_apis.ipynb) | 22.05.2025 |\n| Understanding Language Models | Course deals with language models, in particular (but not exclusively so) on transformer-based language models like GPT-x or LLama | \u003Cul>\u003Cli>[Michael Franke](https:\u002F\u002Fmichael-franke.github.io\u002F)\u003C\u002Fli> \u003Cli>[Carsten Eickhoff](https:\u002F\u002Fgithub.com\u002Feickhoff)\u003C\u002Fli> \u003Cli>[Polina Tsvilodub](https:\u002F\u002Fpolina-tsvilodub.github.io\u002Fhome\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCogSciPrag\u002FUnderstanding-LLMs-course?style=social)](https:\u002F\u002Fgithub.com\u002FCogSciPrag\u002FUnderstanding-LLMs-course) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1301.3781), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2001.08361), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.18223)\u003C\u002Fli>\u003Cli>[book](https:\u002F\u002Fweb.stanford.edu\u002F~jurafsky\u002Fslp3\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fcogsciprag.github.io\u002FUnderstanding-LLMs-course\u002Fintro.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FzjkBMFhNj_g)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FCogSciPrag\u002FUnderstanding-LLMs-course\u002Fblob\u002Fmain\u002Funderstanding-llms\u002Ftutorials\u002F01-introduction.ipynb) | 17.04.2025 |\n| Practical RL | An open course on reinforcement learning in the wild | \u003Cul>\u003Cli>[Pavel Shvechikov](https:\u002F\u002Fgithub.com\u002Fpshvechikov)\u003C\u002Fli> \u003Cli>[Nikita Putintsev](https:\u002F\u002Fgithub.com\u002Fqwasser)\u003C\u002Fli> \u003Cli>[Alexander Fritsler](https:\u002F\u002Fgithub.com\u002FFritz449)\u003C\u002Fli> \u003Cli>[Oleg Vasilev](https:\u002F\u002Fme.svin.in\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dmitry Nikulin](https:\u002F\u002Fgithub.com\u002Fpastafarianist)\u003C\u002Fli> \u003Cli>[Mikhail Konobeev](https:\u002F\u002Fgithub.com\u002Fmknbv)\u003C\u002Fli> \u003Cli>[Ivan Kharitonov](https:\u002F\u002Fkharitonov-ivan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ravil Khisamov](https:\u002F\u002Fgithub.com\u002Fzshrav)\u003C\u002Fli> \u003Cli>[Anna Klepova](https:\u002F\u002Fgithub.com\u002Fq0o0p)\u003C\u002Fli> \u003Cli>[Fedor Ratnikov](https:\u002F\u002Fgithub.com\u002Fjustheuristic)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyandexdataschool\u002FPractical_RL?style=social)](https:\u002F\u002Fgithub.com\u002Fyandexdataschool\u002FPractical_RL) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1703.03864)\u003C\u002Fli>\u003Cli>[feedback](https:\u002F\u002Fdocs.google.com\u002Fforms\u002Fd\u002Fe\u002F1FAIpQLSdurWw97Sm9xCyYwC8g3iB5EibITnoPJW2IkOVQYE_kcXPh6Q\u002Fviewform)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFarama-Foundation\u002FGymnasium)\u003C\u002Fli>\u003Cli>[slides](https:\u002F\u002Fyadi.sk\u002Fd\u002FloPpY45J3EAYfU)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2pWv7GOvuf0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FaUrX-rP_ss4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FlfHX2hHRMVQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FejxfTy4lI6I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FzwYV11a__HQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FD58nLNLkb0I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FxpyKmjJuqhk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FKWlFFtmxjeg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FASQJTNAYdcA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyandexdataschool\u002FPractical_RL\u002Fblob\u002Fmaster\u002Fweek01_intro\u002Fdeep_crossentropy_method.ipynb) | 02.03.2025 |\n| Deep Learning School course (ML + CV) |  | [Nina Konovalova](https:\u002F\u002Fgithub.com\u002FNina-Konovalova) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FDeepLearningSchool\u002Fpart_1_ml_cv?style=social)](https:\u002F\u002Fgithub.com\u002FDeepLearningSchool\u002Fpart_1_ml_cv) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fdatasets\u002Fabrambeyer\u002Fopenintro-possum)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fdls.samcs.ru\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL0Ks75aof3TiHbkJ95vxNlQefujrj1N2w), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL5-da3qGB5ICCsgW1MxlZ0Hq8LL5U3u9y)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FDeepLearningSchool\u002Fpart_1_ml_cv\u002Fblob\u002Fmain\u002Fweek_01_ml_intro\u002FPractice\u002Fpart_1_data_analysis.ipynb) | 14.02.2025 |\n| Introduction to Deep Learning course |  | [Tatiana Gaintseva](https:\u002F\u002Fatmyre.github.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Faifoundation-edu\u002FDL_intro?style=social)](https:\u002F\u002Fgithub.com\u002Faifoundation-edu\u002FDL_intro) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@zxr.nju\u002Fwhat-is-the-kernel-trick-why-is-it-important-98a98db0961d), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fanalytics-vidhya\u002Fhow-relu-works-f317a947bdc6)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fplayground.tensorflow.org\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FKernel_method)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL0Ks75aof3ThuLLtLIVl_KPUDDQlTDyJI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Foa3IVkENrdM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FqvauAflpnyo), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQ7vT0--5VII)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Faifoundation-edu\u002FDL_intro\u002Fblob\u002Fmain\u002F01_Intro_to_Neural_Networks\u002F01_NN_intro.ipynb) | 24.01.2025 |\n| ARENA | Provide talented individuals with the skills, tools, and environment necessary for upskilling in ML engineering, for the purpose of contributing directly to AI alignment in technical roles | [Callum McDougall](https:\u002F\u002Fwww.perfectlynormal.co.uk\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcallummcdougall\u002FARENA_3.0?style=social)](https:\u002F\u002Fgithub.com\u002Fcallummcdougall\u002FARENA_3.0) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.00593)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fslack.svg\" alt=\"slack\" height=20\u002F>](https:\u002F\u002Fjoin.slack.com\u002Ft\u002Farena-uk\u002Fshared_invite\u002Fzt-2noug8mpy-TRYbCnc3pzj7ITNrZIjKww)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Farena-resources.notion.site\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1vuQOB2Gd7OcfzH2y9djXm9OdZA_DcxYz) | 30.12.2024 |\n| Deep Learning Course at the University of Amsterdam | Series of Jupyter notebooks that are designed to help you understanding the \"theory\" from the lectures by seeing corresponding implementations | \u003Cul>\u003Cli>[Pascal Mettes](https:\u002F\u002Fstaff.fnwi.uva.nl\u002Fp.s.m.mettes\u002F)\u003C\u002Fli> \u003Cli>[Melika Davood Zadeh](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FMelika-Davood-Zadeh)\u003C\u002Fli> \u003Cli>[Mohammadreza Salehidehnavi](https:\u002F\u002Fsmsd75.github.io\u002F)\u003C\u002Fli> \u003Cli>[Danilo de Goede](https:\u002F\u002Fivi.fnwi.uva.nl\u002Fvislab\u002Fauthor\u002Fdanilo-de-goede\u002F)\u003C\u002Fli> \u003Cli>[Phillip Lippe](https:\u002F\u002Fphlippe.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fphlippe\u002Fuvadlc_notebooks?style=social)](https:\u002F\u002Fgithub.com\u002Fphlippe\u002Fuvadlc_notebooks) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fuvadlc-notebooks.readthedocs.io\u002Fen\u002Flatest\u002Findex.html)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fuvadlc.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLdlPlO1QhMiAkedeu0aJixfkknLRxk1nA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fphlippe\u002Fuvadlc_notebooks\u002Fblob\u002Fmaster\u002Fdocs\u002Ftutorial_notebooks\u002Ftutorial2\u002FIntroduction_to_PyTorch.ipynb) | 17.10.2024 |\n| The Autodiff Cookbook | You'll go through a whole bunch of neat autodiff ideas that you can cherry pick for your own work, starting with the basics | \u003Cul>\u003Cli>[Alex Wiltschko](https:\u002F\u002Fgithub.com\u002Falexbw)\u003C\u002Fli> \u003Cli>[Matthew Johnson](http:\u002F\u002Fpeople.csail.mit.edu\u002Fmattjj\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fjax?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fjax\u002Fissues\u002F446#issuecomment-467105048) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1406.2572), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.04454), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.03451), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.07062)\u003C\u002Fli>\u003Cli>[book](https:\u002F\u002Fmitpress.mit.edu\u002Fsites\u002Fdefault\u002Ffiles\u002Ftitles\u002Fcontent\u002Fsicm_edition_2\u002Fbook.html), [book](https:\u002F\u002Fmitpress.mit.edu\u002Fbooks\u002Ffunctional-differential-geometry)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fjax#auto-vectorization-with-vmap), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhips\u002Fautograd)\u003C\u002Fli>\u003Cli>[tutorial](http:\u002F\u002Fvideolectures.net\u002Fdeeplearning2017_johnson_automatic_differentiation\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FTruncated_Newton_method), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FPullback_(differential_geometry)), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FHolomorphic_function), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FCauchy%E2%80%93Riemann_equations)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fjax\u002Fblob\u002Fmain\u002Fdocs\u002Fnotebooks\u002Fautodiff_cookbook.ipynb) | 20.09.2024 |\n| Machine Learning Simplified | A Gentle Introduction to Supervised Learning | [Andrew Wolf](https:\u002F\u002F5x12.ai\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002F5x12\u002Fthemlsbook?style=social)](https:\u002F\u002Fgithub.com\u002F5x12\u002Fthemlsbook) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fgeekculture\u002Fi-found-a-great-machine-learning-book-deed11db2688)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FPython\u002Fcomments\u002Ft8st9l\u002Fi_wrote_a_book_on_machine_learning_w_python_code\u002F), [\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Flearnmachinelearning\u002Fcomments\u002Fsnxlly\u002Fmachine_learning_simplified_book\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fwww.themlsbook.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002F5x12\u002Fthemlsbook\u002Fblob\u002Fmaster\u002Fchapter2\u002Fknn.ipynb) | 29.08.2024 |\n| Anthropic courses | Anthropic's educational courses | [Anthropic](https:\u002F\u002Fwww.anthropic.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fanthropics\u002Fcourses?style=social)](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fcourses) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fdocs.anthropic.com\u002Fen\u002Fdocs\u002Fresources\u002Fcourses)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fanthropic-sdk-python)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fanthropic\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FClaudeAI\u002Fcomments\u002F1f7czsx\u002Fanthropics_official_educational_courses_on_prompt\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fanthropics\u002Fcourses\u002Fblob\u002Fmaster\u002Fanthropic_api_fundamentals\u002F01_getting_started.ipynb) | 22.08.2024 |\n| mlcourse.ai | Open Machine Learning Course | [Yury Kashnitsky](https:\u002F\u002Fyorko.github.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYorko\u002Fmlcourse.ai?style=social)](https:\u002F\u002Fgithub.com\u002FYorko\u002Fmlcourse.ai) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fhabr.com\u002Fcompany\u002Fods\u002Fblog\u002F344044\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fkashnitsky\u002Fmlcourse)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fopen-machine-learning-course)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmlcourse.ai\u002Fbook\u002Findex.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fslack.svg\" alt=\"slack\" height=20\u002F>](https:\u002F\u002Fopendatascience.slack.com\u002Farchives\u002FC91N8TL83\u002Fp1567408586359500)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLVlY_7IJCMJeRfZ68eVfEcu-UcN9BbwiX)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FYorko\u002Fmlcourse.ai\u002Fblob\u002Fmain\u002Fjupyter_english\u002Ftopic01_pandas_data_analysis\u002Ftopic1_pandas_data_analysis.ipynb) | 19.08.2024 |\n| Deep RL Course | The Hugging Face Deep Reinforcement Learning Course | \u003Cul>\u003Cli>[Thomas Simonini](https:\u002F\u002Fwww.simoninithomas.com\u002F)\u003C\u002Fli> \u003Cli>[Omar Sanseviero](https:\u002F\u002Fosanseviero.github.io\u002Fhackerllama\u002F)\u003C\u002Fli> \u003Cli>[Sayak Paul](https:\u002F\u002Fsayak.dev\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhuggingface\u002Fdeep-rl-class?style=social)](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fdeep-rl-class) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FydHrjt3WP5)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Falex-petrenko\u002Fsample-factory)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdeep-rl-course\u002Funit0\u002Fintroduction), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fhuggingface-projects\u002FDeep-Reinforcement-Learning-Leaderboard)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fpytorch.org\u002Ftutorials\u002Fbeginner\u002Fdeep_learning_60min_blitz.html)\u003C\u002Fli>\u003Cli>[syllabus](https:\u002F\u002Fsimoninithomas.github.io\u002Fdeep-rl-course)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2GwBez0D20A), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FCsuIANBnSq8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAQKAOXJa6qg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fhuggingface\u002Fdeep-rl-class\u002Fblob\u002Fmain\u002Fnotebooks\u002Funit1\u002Funit1.ipynb) | 24.06.2024 |\n| Anthropic's Prompt Engineering Interactive Tutorial | Course is intended to provide you with a comprehensive step-by-step understanding of how to engineer optimal prompts within Claude | [Anthropic](https:\u002F\u002Fwww.anthropic.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fanthropics\u002Fprompt-eng-interactive-tutorial?style=social)](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fprompt-eng-interactive-tutorial) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fdocs.anthropic.com\u002Fen\u002Fdocs\u002Fbuild-with-claude\u002Fprompt-engineering\u002Foverview)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fanthropic-sdk-python)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fcoding-nexus\u002Fmastering-prompt-engineering-for-ai-a-free-hands-on-guide-with-anthropics-tutorial-6f2cabcbde5a)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FT9aRN5JkmL8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhkhDdcM5V94)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fanthropics\u002Fprompt-eng-interactive-tutorial\u002Fblob\u002Fmaster\u002FAnthropic%201P\u002F00_Tutorial_How-To.ipynb) | 02.04.2024 |\n| Generative AI for Beginners - A Course | A 12 Lesson course teaching everything you need to know to start building Generative AI applications | [microsoft](https:\u002F\u002Fwww.microsoft.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002Fxr-development-for-beginners?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fxr-development-for-beginners) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Faka.ms\u002Fgenai-discord)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FWeb-Dev-For-Beginners)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmicrosoft.github.io\u002Fgenerative-ai-for-beginners\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmicrosoft\u002Fgenerative-ai-for-beginners\u002Fblob\u002Fmain\u002F06-text-generation-apps\u002Fnotebook-azure-openai.ipynb) | 22.02.2024 |\n| Deep Reinforcement Learning | CS 285 at UC Berkeley | \u003Cul>\u003Cli>[Sergey Levine](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~svlevine\u002F)\u003C\u002Fli> \u003Cli>[Kyle Stachowicz](https:\u002F\u002Fkylestaschowicz.com)\u003C\u002Fli> \u003Cli>[Vivek Myers](https:\u002F\u002Fvmyers.github.io)\u003C\u002Fli> \u003Cli>[Joey Hong](https:\u002F\u002Fjoeyhong123.github.io)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Kevin Black](https:\u002F\u002Fkvablack.github.io)\u003C\u002Fli> \u003Cli>[Michael Janner](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~janner\u002F)\u003C\u002Fli> \u003Cli>[Vitchyr Pong](https:\u002F\u002Fvitchyr.github.io)\u003C\u002Fli> \u003Cli>[Aviral Kumar](https:\u002F\u002Faviralkumar2907.github.io)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fberkeleydeeprlcourse\u002Fhomework_fall2023?style=social)](https:\u002F\u002Fgithub.com\u002Fberkeleydeeprlcourse\u002Fhomework_fall2023) \u003Cul>\u003Cli>[coursera](https:\u002F\u002Fwww.coursera.org\u002Fcourse\u002Fneuralnets), [coursera](https:\u002F\u002Fwww.coursera.org\u002Flearn\u002Fmachine-learning\u002F)\u003C\u002Fli>\u003Cli>[website](http:\u002F\u002Frail.eecs.berkeley.edu\u002Fdeeprlcourse\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL_iWQOsE6TfVYGEGiAOMaOzzv41Jfm_Ps), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutube.com\u002Fplaylist?list=PL_iWQOsE6TfX7MaC6C3HcdOf1g337dlC9)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fberkeleydeeprlcourse\u002Fhomework_fall2023\u002Fblob\u002Fmaster\u002Fhw1\u002Fcs285\u002Fscripts\u002Frun_hw1.ipynb) | 29.08.2023 |\n| npNLG | The course introduces the basics of NLG, neural language models and their implementation in PyTorch, as well as a selection of recent pragmatic neural NLG approaches | [Michael Franke](https:\u002F\u002Fgithub.com\u002Fmichael-franke) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a1b5a7212442.png)](https:\u002F\u002Fdoi.org\u002F10.1126\u002Fscience.1218633) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmichael-franke\u002FnpNLG?style=social)](https:\u002F\u002Fgithub.com\u002Fmichael-franke\u002FnpNLG) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.09867)\u003C\u002Fli>\u003Cli>[book](https:\u002F\u002Fmichael-franke.github.io\u002FnpNLG\u002F000-intro.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmichael-franke\u002FnpNLG\u002Fblob\u002Fmain\u002Fneural_pragmatic_nlg\u002F01-prob-prag\u002F01b-RSA-vanilla.ipynb) | 09.11.2022 |\n| DSP theory | Theory of digital signal processing: signals, filtration (IIR, FIR, CIC, MAF), transforms (FFT, DFT, Hilbert, Z-transform) etc | \u003Cul>\u003Cli>[Alexander Kapitanov](https:\u002F\u002Fgithub.com\u002Fhukenovs)\u003C\u002Fli> \u003Cli>[Vladimir Fadeev](https:\u002F\u002Fgithub.com\u002Fkirlf)\u003C\u002Fli> \u003Cli>[Karina Kvanchiani](https:\u002F\u002Fgithub.com\u002Fkarinakvanchiani)\u003C\u002Fli> \u003Cli>[Elizaveta Petrova](https:\u002F\u002Fgithub.com\u002Fkleinsbotle)\u003C\u002Fli> \u003Cli>[Andrei Makhliarchuk](https:\u002F\u002Fgithub.com\u002Fanotherhelloworld)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhukenovs\u002Fdsp-theory?style=social)](https:\u002F\u002Fgithub.com\u002Fhukenovs\u002Fdsp-theory) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fhabr.com\u002Fru\u002Farticles\u002F460445\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fhukenovs\u002Fdsp-theory\u002Fblob\u002Fmaster\u002Fsrc\u002Fdsp_theory_1_signals.ipynb) | 18.10.2022 |\n| Machine learning course | This course is broad and shallow, but author will provide additional links so that you can deepen your understanding of the ML method you need | [Тимчишин Віталій](https:\u002F\u002Fgithub.com\u002Ffbeilstein) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffbeilstein\u002Fmachine_learning?style=social)](https:\u002F\u002Fgithub.com\u002Ffbeilstein\u002Fmachine_learning) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fvas3k.com\u002Fblog\u002Fmachine_learning\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLkDeTjsoxDVgnb2lIYo9-1l4XYhrIyS6A), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-RdOwhmqP5s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FR13BD8qKeTg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FZkjP5RJLQF4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJ4Wdy0Wc_xQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FmBcLRGuAFUk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYIGtalP1mv0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYz5pySyEtsU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fx5zLaWT5KPs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyBwpo-L80Mc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL3FW7Lu3i5JvHM8ljYj-zLfQRF3EO8sYv)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffbeilstein\u002Fmachine_learning\u002Fblob\u002Fmaster\u002Flecture_01_introduction.ipynb) | 02.09.2021 |\n| Udacity Deep Learning class with TensorFlow | Learn how to apply deep learning to solve complex problems | [Mark Daoust](https:\u002F\u002Fgithub.com\u002FMarkDaoust) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Fexamples?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fexamples\u002Ftree\u002Fmaster\u002Fcourses\u002Fudacity_deep_learning) \u003Cul>\u003Cli>[data](http:\u002F\u002Fyaroslavvb.blogspot.com\u002F2011\u002F09\u002Fnotmnist-dataset.html), [data](http:\u002F\u002Fyann.lecun.com\u002Fexdb\u002Fmnist\u002F)\u003C\u002Fli>\u003Cli>[udacity](https:\u002F\u002Fwww.udacity.com\u002Fcourse\u002Fdeep-learning--ud730)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLAwxTw4SYaPn_OWPFT9ulXLuQrImzHfOV)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fexamples\u002Fblob\u002Fmaster\u002Fcourses\u002Fudacity_deep_learning\u002F1_notmnist.ipynb) | 20.01.2021 |\n| Intro to TensorFlow for Deep Learning | Dive into deep learning with this practical course on TensorFlow and the Keras API | \u003Cul>\u003Cli>[Magnus Hyttsten](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fmagnushyttsten)\u003C\u002Fli> \u003Cli>[Juan Delgado](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fjuan-delgado-845749179)\u003C\u002Fli> \u003Cli>[Paige Bailey](https:\u002F\u002Fwebpaige.dev\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Fexamples?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fexamples\u002Ftree\u002Fmaster\u002Fcourses\u002Fudacity_intro_to_tensorflow_for_deep_learning) \u003Cul>\u003Cli>[udacity](https:\u002F\u002Fwww.udacity.com\u002Fcourse\u002Fintro-to-tensorflow-for-deep-learning--ud187)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fexamples\u002Fblob\u002Fmaster\u002Fcourses\u002Fudacity_intro_to_tensorflow_for_deep_learning\u002Fl02c01_celsius_to_fahrenheit.ipynb) | 09.09.2020 |\n| Introduction to TensorFlow Lite | Learn how to deploy deep learning models on mobile and embedded devices with TensorFlow Lite | \u003Cul>\u003Cli>[Daniel Situnayake](https:\u002F\u002Fsitunayake.com\u002F)\u003C\u002Fli> \u003Cli>[Paige Bailey](https:\u002F\u002Fwebpaige.dev\u002F)\u003C\u002Fli> \u003Cli>[Juan Delgado](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fjuan-delgado-845749179)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Fexamples?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fexamples\u002Ftree\u002Fmaster\u002Fcourses\u002Fudacity_intro_to_tensorflow_lite) \u003Cul>\u003Cli>[udacity](https:\u002F\u002Fwww.udacity.com\u002Fcourse\u002Fintro-to-tensorflow-lite--ud190)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fexamples\u002Fblob\u002Fmaster\u002Fcourses\u002Fudacity_intro_to_tensorflow_lite\u002Ftflite_c01_linear_regression.ipynb) | 09.09.2020 |\n| NYU-DLSP20 | This course concerns the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition | \u003Cul>\u003Cli>[Yann LeCun](https:\u002F\u002Fyann.lecun.com\u002F)\u003C\u002Fli> \u003Cli>[Alfredo Canziani](https:\u002F\u002Fatcold.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAtcold\u002FNYU-DLSP20?style=social)](https:\u002F\u002Fgithub.com\u002FAtcold\u002FNYU-DLSP20) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FCthuqsX8Pb)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAtcold\u002FNYU-DLSP21), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAtcold\u002FNYU-DLFL22)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FNYU_DeepLearning\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fatcold.github.io\u002FNYU-DLSP20\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLLHTzKZzVU9eaEyErdV26ikyolxOsz6mq)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAtcold\u002FNYU-DLSP20\u002Fblob\u002Fmaster\u002F00-logic_neuron_programming.ipynb) | 30.10.2019 |\n\n\u003C\u002Fdetails>\n\n## Research\n\u003Cdetails>\n\u003Csummary>RESEARCH\u003C\u002Fsummary>\n\n| name | description | authors | links | colaboratory | update |\n|------|-------------|:--------|:------|:------------:|:------:|\n| GigaAM | SSL pretraining framework that leverages masked language modeling with targets derived from a speech recognition model | \u003Cul>\u003Cli>[Aleksandr Kutsakov](https:\u002F\u002Fgithub.com\u002FAlexander4127)\u003C\u002Fli> \u003Cli>[Alexandr Maximenko](https:\u002F\u002Fgithub.com\u002FAlexMaximenko)\u003C\u002Fli> \u003Cli>[Georgii Gospodinov](https:\u002F\u002Fgithub.com\u002Fgeorgygospodinov)\u003C\u002Fli> \u003Cli>[Pavel Bogomolov](https:\u002F\u002Fgithub.com\u002FBobrosoft98)\u003C\u002Fli> \u003Cli>[Fyodor Minkin](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FFyodor-Minkin)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsalute-developers\u002FGigaAM?style=social)](https:\u002F\u002Fgithub.com\u002Fsalute-developers\u002FGigaAM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.01192), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.08100), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1211.3711), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.11477), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.01192)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fhabr.com\u002Fru\u002Fcompanies\u002Fsberdevices\u002Farticles\u002F805569)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fai-sage\u002FGigaAM-v3), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fai-sage\u002FGigaAM-v3)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgigaam\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FO7NSH2SAwRc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FPvZuTUnZa2Q), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FktO4Mx6UMNk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F7NEvFNtwRTA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsalute-developers\u002FGigaAM\u002Fblob\u002Fmain\u002Fcolab_example.ipynb) | 20.11.2025 |\n| Segment Anything 3 | Unified model that detects, segments, and tracks objects in images and videos based on concept prompts, which we define as either short noun phrases (e.g., “yellow school bus”), image exemplars, or a combination of both | \u003Cul>\u003Cli>[Nicolas Carion](https:\u002F\u002Fwww.nicolascarion.com\u002F)\u003C\u002Fli> \u003Cli>[Laura Gustafson](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=c8IpF9gAAAAJ)\u003C\u002Fli> \u003Cli>[Yuan-Ting Hu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=E8DVVYQAAAAJ)\u003C\u002Fli> \u003Cli>[Shoubhik Debnath](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=fb6FOfsAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ronghang Hu](https:\u002F\u002Fronghanghu.com\u002F)\u003C\u002Fli> \u003Cli>[Didac Suris](https:\u002F\u002Fwww.didacsuris.com\u002F)\u003C\u002Fli> \u003Cli>[Chaitanya Ryali](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4LWx24UAAAAJ)\u003C\u002Fli> \u003Cli>[Kalyan Vasudev Alwala](https:\u002F\u002Fscholar.google.co.in\u002Fcitations?user=m34oaWEAAAAJ)\u003C\u002Fli> \u003Cli>[Haitham Khedr](https:\u002F\u002Fhkhedr.com\u002F)\u003C\u002Fli> \u003Cli>[Andrew Huang]()\u003C\u002Fli> \u003Cli>[Jie Lei](https:\u002F\u002Fjayleicn.github.io)\u003C\u002Fli> \u003Cli>[Tengyu Ma](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=VeTSl0wAAAAJ)\u003C\u002Fli> \u003Cli>[Baishan Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=BC5wDu8AAAAJ)\u003C\u002Fli> \u003Cli>[Arpit Kalla](https:\u002F\u002Fgithub.com\u002Farpitkalla)\u003C\u002Fli> \u003Cli>[Markus Marks](https:\u002F\u002Fdamaggu.github.io)\u003C\u002Fli> \u003Cli>[Joseph Greer](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=guL96CkAAAAJ)\u003C\u002Fli> \u003Cli>[Meng Wang]()\u003C\u002Fli> \u003Cli>[Peize Sun](https:\u002F\u002Fpeizesun.github.io)\u003C\u002Fli> \u003Cli>[Roman Rädle](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Tpt57v0AAAAJ)\u003C\u002Fli> \u003Cli>[Triantafyllos Afouras](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~afourast)\u003C\u002Fli> \u003Cli>[Effrosyni Mavroudi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vYRzGGEAAAAJ)\u003C\u002Fli> \u003Cli>[Katherine Xu](https:\u002F\u002Fk8xu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tsung-Han Wu](https:\u002F\u002Fpatrickthwu.com\u002F)\u003C\u002Fli> \u003Cli>[Yu Zhou](https:\u002F\u002Fyu-bryan-zhou.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liliane Momeni](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Lb-KgVYAAAAJ)\u003C\u002Fli> \u003Cli>[Rishi Hazra](https:\u002F\u002Frishihazra.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shuangrui Ding](https:\u002F\u002Fmark12ding.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sagar Vaze](https:\u002F\u002Fsgvaze.github.io\u002F)\u003C\u002Fli> \u003Cli>[Francois Porcher](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=LgHZ8hUAAAAJ)\u003C\u002Fli> \u003Cli>[Feng Li](https:\u002F\u002Ffengli-ust.github.io\u002F)\u003C\u002Fli> \u003Cli>[Siyuan Li](https:\u002F\u002Fsiyuanliii.github.io\u002F)\u003C\u002Fli> \u003Cli>[Aishwarya Kamath](https:\u002F\u002Fashkamath.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ho Kei Cheng](https:\u002F\u002Fhkchengrex.com\u002F)\u003C\u002Fli> \u003Cli>[Piotr Dollar](https:\u002F\u002Fpdollar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nikhila Ravi](https:\u002F\u002Fnikhilaravi.com\u002F)\u003C\u002Fli> \u003Cli>[Kate Saenko](https:\u002F\u002Fai.bu.edu\u002Fksaenko.html)\u003C\u002Fli> \u003Cli>[Pengchuan Zhang](https:\u002F\u002Fpzzhang.github.io\u002Fpzzhang\u002F)\u003C\u002Fli> \u003Cli>[Christoph Feichtenhofer](https:\u002F\u002Ffeichtenhofer.github.io)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fsam3?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fsam3) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.16719)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Faidemos.meta.com\u002Fsegment-anything)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002Fsam3), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Ffacebook\u002FSACo-Gold), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Ffacebook\u002FSACo-Silver), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Ffacebook\u002FSACo-VEval)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-sam-3-segment-anything-with-concepts-18e9501df00e)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresearch\u002Fpublications\u002Fsam-3-segment-anything-with-concepts\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fsam3\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fsegment-anything-model-3\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fabout.fb.com\u002Fnews\u002F2025\u002F11\u002Fnew-sam-models-detect-objects-create-3d-reconstructions\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fsam3\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002F1p1cfvx\u002Fr_segment_anything_model_3_sam_3_is_released\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLIsFyaUd-B30iOHlOHOb3OX6V_dum_DtN), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIATrWxEDpu4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fsam3\u002Fblob\u002Fmain\u002Fexamples\u002Fsam3_for_sam2_video_task_example.ipynb) | 19.11.2025 |\n| AlphaFold | Highly accurate protein structure prediction | \u003Cul>\u003Cli>[John Jumper](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=a5goOh8AAAAJ)\u003C\u002Fli> \u003Cli>[Richard Evans](http:\u002F\u002Fwww.doc.ic.ac.uk\u002F~re14\u002F)\u003C\u002Fli> \u003Cli>[Alexander Pritzel](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=GPgAyU0AAAAJ)\u003C\u002Fli> \u003Cli>[Tim Green](http:\u002F\u002Ftfgg.me\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Michael Figurnov](https:\u002F\u002Ffigurnov.ru\u002F)\u003C\u002Fli> \u003Cli>[Olaf Ronneberger](https:\u002F\u002Flmb.informatik.uni-freiburg.de\u002Fpeople\u002Fronneber\u002F)\u003C\u002Fli> \u003Cli>[Kathryn Tunyasuvunakool](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=eEqNGagAAAAJ)\u003C\u002Fli> \u003Cli>[Russ Bates](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Koes5ewAAAAJ)\u003C\u002Fli> \u003Cli>[Augustin Žídek](https:\u002F\u002Faugustin.zidek.eu\u002F)\u003C\u002Fli> \u003Cli>[Anna Potapenko](http:\u002F\u002Fapotapenko.com\u002F)\u003C\u002Fli> \u003Cli>[Alex Bridgland](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=VWmXKPMAAAAJ)\u003C\u002Fli> \u003Cli>[Clemens Meyer](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=EWLZiM8AAAAJ)\u003C\u002Fli> \u003Cli>[Simon Kohl](https:\u002F\u002Fwww.simonkohl.com\u002F)\u003C\u002Fli> \u003Cli>[Andrew Ballard](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=syjQhAMAAAAJ)\u003C\u002Fli> \u003Cli>[Bernardino Romera-Paredes](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fromeraparedes\u002F)\u003C\u002Fli> \u003Cli>[Stanislav Nikolov](https:\u002F\u002Fscholar.google.co.uk\u002Fcitations?user=O-b7pBEAAAAJ)\u003C\u002Fli> \u003Cli>[Rishub Jain](http:\u002F\u002Frishub.me\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_80dd955e0b2f.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41586-021-03819-2) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepmind\u002Falphafold?style=social)](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Falphafold) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fdeepmind.com\u002Fblog\u002Farticle\u002Falphafold-a-solution-to-a-50-year-old-grand-challenge-in-biology), [blog post](https:\u002F\u002Fdeepmind.com\u002Fblog\u002Farticle\u002Fputting-the-power-of-alphafold-into-the-worlds-hands)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsokrypton\u002FColabFold), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Ftree), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fchex)\u003C\u002Fli>\u003Cli>[paper](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-021-03828-1)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fmethod\u002Falphafold)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FAlphaFold)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=gg7WjuFs8F4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=B9PL__gVxLI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsokrypton\u002FColabFold\u002Fblob\u002Fmain\u002FAlphaFold2.ipynb) | 23.10.2025 |\n| OWL-ViT | Simple Open-Vocabulary Object Detection with Vision Transformers | \u003Cul>\u003Cli>[Matthias Minderer](http:\u002F\u002Fmatthias.minderer.net\u002F)\u003C\u002Fli> \u003Cli>[Alexey Gritsenko](https:\u002F\u002Fgithub.com\u002FAlexeyG)\u003C\u002Fli> \u003Cli>[Austin Stone](https:\u002F\u002Fgithub.com\u002FAustinCStone)\u003C\u002Fli> \u003Cli>[Maxim Neumann](https:\u002F\u002Fgithub.com\u002Fmaximneumann)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dirk Weissenborn](https:\u002F\u002Fgithub.com\u002Fdirkweissenborn)\u003C\u002Fli> \u003Cli>[Alexey Dosovitskiy](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=FXNJRDoAAAAJ)\u003C\u002Fli> \u003Cli>[Aravindh Mahendran](https:\u002F\u002Fgithub.com\u002Faravindhm)\u003C\u002Fli> \u003Cli>[Anurag Arnab](https:\u002F\u002Fgithub.com\u002Fanuragarnab)\u003C\u002Fli> \u003Cli>[Mostafa Dehghani](https:\u002F\u002Fmostafadehghani.com\u002F)\u003C\u002Fli> \u003Cli>[Zhuoran Shen](https:\u002F\u002Fcmsflash.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xiao Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ukyXqzMAAAAJ)\u003C\u002Fli> \u003Cli>[Xiaohua Zhai](https:\u002F\u002Fgithub.com\u002Fxiaohuazhai)\u003C\u002Fli> \u003Cli>[Thomas Kipf](https:\u002F\u002Ftkipf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Neil Houlsby](https:\u002F\u002Fneilhoulsby.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9d15beeb274f.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20080-9_42) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.06230)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Fowlvit)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fhuggingface\u002Fnotebooks\u002Fblob\u002Fmain\u002Fexamples\u002Fzeroshot_object_detection_with_owlvit.ipynb) | 29.09.2025 |\n| CWM | Code World Model, a 32-billion-parameter open-weights LLM, to advance research on code generation with world models | \u003Cul>\u003Cli>[Jade Copet](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=GRMLwjAAAAAJ)\u003C\u002Fli> \u003Cli>[Quentin Carbonneaux](https:\u002F\u002Ffr.linkedin.com\u002Fin\u002Fquentin-carbonneaux-36730717a)\u003C\u002Fli> \u003Cli>[Gal Cohen](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fgal-cohen-29b980122)\u003C\u002Fli> \u003Cli>[Jonas Gehring](https:\u002F\u002Fjgehring.net)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jacob Kahn](https:\u002F\u002Fjacobkahn.me)\u003C\u002Fli> \u003Cli>[Jannik Kossen](https:\u002F\u002Fopenreview.net\u002Fprofile?id=~Jannik_Kossen2)\u003C\u002Fli> \u003Cli>[Felix Kreuk](https:\u002F\u002Fopenreview.net\u002Fprofile?id=~Felix_Kreuk1)\u003C\u002Fli> \u003Cli>[Emily McMilin](https:\u002F\u002F2dot71mily.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michel Meyer](https:\u002F\u002Fwww.frenchweb.fr\u002Fa-la-rencontre-de-michel-meyer-technical-program-manager-pour-facebook-base-a-san-francisco\u002F427606)\u003C\u002Fli> \u003Cli>[Yuxiang Wei](https:\u002F\u002Fyuxiang.cs.illinois.edu)\u003C\u002Fli> \u003Cli>[David Zhang](https:\u002F\u002Fdavzha.netlify.app)\u003C\u002Fli> \u003Cli>[Kunhao Zheng](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=zDy4jSYAAAAJ)\u003C\u002Fli> \u003Cli>[Jordi Armengol Estape](http:\u002F\u002Fjordiae.com)\u003C\u002Fli> \u003Cli>[Pedram Bashiri](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=weIwEtgAAAAJ)\u003C\u002Fli> \u003Cli>[Maximilian Beck](https:\u002F\u002Fmaximilianmbeck.github.io)\u003C\u002Fli> \u003Cli>[Pierre Chambon](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5e_2WW4AAAAJ)\u003C\u002Fli> \u003Cli>[Abhishek Charnalia](https:\u002F\u002Fuk.linkedin.com\u002Fin\u002Fabhishekcharnalia)\u003C\u002Fli> \u003Cli>[Chris Cummins](https:\u002F\u002Fchriscummins.cc)\u003C\u002Fli> \u003Cli>[Juliette Decugis](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QXSA4p4AAAAJ)\u003C\u002Fli> \u003Cli>[Zacharias Fisches](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=nYfsAVcAAAAJ)\u003C\u002Fli> \u003Cli>[François Fleuret](https:\u002F\u002Ffleuret.org)\u003C\u002Fli> \u003Cli>[Fabian Gloeckle](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=_dF0fX4AAAAJ)\u003C\u002Fli> \u003Cli>[Alex Gu](https:\u002F\u002Fminimario.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michael Hassid](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5d93np0AAAAJ)\u003C\u002Fli> \u003Cli>[Daniel Haziza](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2eSKdFMAAAAJ)\u003C\u002Fli> \u003Cli>[Badr Youbi Idrissi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=xB5UivYAAAAJ)\u003C\u002Fli> \u003Cli>[Christian Keller](https:\u002F\u002Ffr.linkedin.com\u002Fin\u002Fkellerchristian)\u003C\u002Fli> \u003Cli>[Rahul Kindi](https:\u002F\u002Fgithub.com\u002Frkindi)\u003C\u002Fli> \u003Cli>[Hugh Leather](https:\u002F\u002Fhomepages.inf.ed.ac.uk\u002Fhleather\u002F)\u003C\u002Fli> \u003Cli>[Gallil Maimon](https:\u002F\u002Fpages.cs.huji.ac.il\u002Fgallilmaimon\u002F)\u003C\u002Fli> \u003Cli>[Aram Markosyan](https:\u002F\u002Fwww.markosyanaram.com\u002F)\u003C\u002Fli> \u003Cli>[Francisco Massa](https:\u002F\u002Ffmassa.github.io)\u003C\u002Fli> \u003Cli>[Pierre-Emmanuel Mazaré](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=IUod9t8AAAAJ)\u003C\u002Fli> \u003Cli>[Vegard Mella](https:\u002F\u002Fgithub.com\u002Ftscmoo)\u003C\u002Fli> \u003Cli>[Naila Murray](https:\u002F\u002Fnailamurray.github.io)\u003C\u002Fli> \u003Cli>[Keyur Muzumdar](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5mkv6-kAAAAJ)\u003C\u002Fli> \u003Cli>[Peter O’Hearn](https:\u002F\u002Fai.meta.com\u002Fpeople\u002F567943484717417\u002Fpeter-ohearn\u002F)\u003C\u002Fli> \u003Cli>[Matteo Pagliardini](https:\u002F\u002Fmatteopagliardini.github.io)\u003C\u002Fli> \u003Cli>[Dmitrii Pedchenko](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=izGGAnQAAAAJ)\u003C\u002Fli> \u003Cli>[Tal Remez](https:\u002F\u002Ftalremez.github.io\u002F)\u003C\u002Fli> \u003Cli>[Volker Seeker](https:\u002F\u002Fwww.volkerseeker.com\u002F)\u003C\u002Fli> \u003Cli>[Marco Selvi](https:\u002F\u002Fgithub.com\u002Fmarcoselvi)\u003C\u002Fli> \u003Cli>[Oren Sultan](https:\u002F\u002Fgithub.com\u002Forensul)\u003C\u002Fli> \u003Cli>[Sida Wang](https:\u002F\u002Fai.meta.com\u002Fpeople\u002F327339230340379\u002Fsida-wang\u002F)\u003C\u002Fli> \u003Cli>[Luca Wehrstedt](https:\u002F\u002Fgithub.com\u002Flw)\u003C\u002Fli> \u003Cli>[Ori Yoran](https:\u002F\u002Fwww.oriyoran.com\u002F)\u003C\u002Fli> \u003Cli>[Lingming Zhang](https:\u002F\u002Flingming.cs.illinois.edu)\u003C\u002Fli> \u003Cli>[Taco Cohen](https:\u002F\u002Ftacocohen.com)\u003C\u002Fli> \u003Cli>[Yossi Adi](https:\u002F\u002Fai.meta.com\u002Fpeople\u002F1930846650664668\u002Fyossi-adi\u002F)\u003C\u002Fli> \u003Cli>[Gabriel Synnaeve](https:\u002F\u002Fai.meta.com\u002Fpeople\u002F1447559096135307\u002Fgabriel-synnaeve\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fcwm?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fcwm) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002Fcwm), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002Fcwm-sft), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002Fcwm-pretrain)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science-in-your-pocket\u002Fmeta-code-world-models-released-f988a2f92e71), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fnoailabs.medium.com\u002Fcode-world-model-the-dawn-of-self-aware-software-b07a37cfd600)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresearch\u002Fpublications\u002Fcwm\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresources\u002Fmodels-and-libraries\u002Fcwm-downloads)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FAIatMeta\u002Fstatus\u002F1970963571753222319)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FvUYhfbj52E0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F9yRxXCJDkS0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fcwm\u002Fblob\u002Fmaster\u002Fdemos\u002Fcwmdbg.ipynb) | 24.09.2025 |\n| Qwen3-Omni | Single multimodal model that, for the first time, maintains state-of-the-art performance across text, image, audio, and video without any degradation relative to single-modal counterparts | \u003Cul>\u003Cli>[Jin Xu](https:\u002F\u002Fjxu-thu.github.io)\u003C\u002Fli> \u003Cli>[Zhifang Guo](https:\u002F\u002Fopenreview.net\u002Fprofile?id=~Zhifang_Guo3)\u003C\u002Fli> \u003Cli>[Hangrui Hu](https:\u002F\u002Fgithub.com\u002Fhangruihu)\u003C\u002Fli> \u003Cli>[Yunfei Chu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=41QhCyYAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xiong Wang](https:\u002F\u002Fgithub.com\u002Fwangxiongts)\u003C\u002Fli> \u003Cli>[Jinzheng He](https:\u002F\u002Fgithub.com\u002FJinzhengHe)\u003C\u002Fli> \u003Cli>[Yuxuan Wang](https:\u002F\u002Fyuxuanw.me)\u003C\u002Fli> \u003Cli>[Xian Shi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=UCiUsSEAAAAJ)\u003C\u002Fli> \u003Cli>[Ting He](https:\u002F\u002Fgithub.com\u002Ftinghe)\u003C\u002Fli> \u003Cli>[Xinfa Zhu](https:\u002F\u002Fgithub.com\u002FXinfaZhu)\u003C\u002Fli> \u003Cli>[Yuanjun Lv](https:\u002F\u002Fgithub.com\u002FYuanjunLv)\u003C\u002Fli> \u003Cli>[Yongqi Wang](https:\u002F\u002Fgithub.com\u002FYongqiWang03)\u003C\u002Fli> \u003Cli>[Dake Guo](https:\u002F\u002Fgithub.com\u002FDakeGuo)\u003C\u002Fli> \u003Cli>[He Wang](https:\u002F\u002Fgithub.com\u002Fhwwang21)\u003C\u002Fli> \u003Cli>[Linhan Ma](https:\u002F\u002Fgithub.com\u002FLinhanMa)\u003C\u002Fli> \u003Cli>[Pei Zhang](https:\u002F\u002Fgithub.com\u002FPeiZhang98)\u003C\u002Fli> \u003Cli>[Xinyu Zhang](https:\u002F\u002Fgithub.com\u002FXinyuZhang97)\u003C\u002Fli> \u003Cli>[Hongkun Hao](https:\u002F\u002Fgithub.com\u002FHongkunHao)\u003C\u002Fli> \u003Cli>[Zishan Guo](https:\u002F\u002Fgithub.com\u002FZishanGuo)\u003C\u002Fli> \u003Cli>[Baosong Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=DE68IOIAAAAJ)\u003C\u002Fli> \u003Cli>[Bin Zhang](https:\u002F\u002Fgithub.com\u002Fbinzhaozhao)\u003C\u002Fli> \u003Cli>[Ziyang Ma](https:\u002F\u002Fgithub.com\u002FZiyangMa)\u003C\u002Fli> \u003Cli>[Xipin Wei](https:\u002F\u002Fgithub.com\u002FXipinWei)\u003C\u002Fli> \u003Cli>[Shuai Bai](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=iv2dcv8AAAAJ)\u003C\u002Fli> \u003Cli>[Keqin Chen](https:\u002F\u002Fgithub.com\u002Fkeqinc)\u003C\u002Fli> \u003Cli>[Xuejing Liu](https:\u002F\u002Fgithub.com\u002FXuejingL)\u003C\u002Fli> \u003Cli>[Peng Wang](https:\u002F\u002Fgithub.com\u002Fpengwang114)\u003C\u002Fli> \u003Cli>[Mingkun Yang](https:\u002F\u002Fgithub.com\u002FMingkunYang)\u003C\u002Fli> \u003Cli>[Dayiheng Liu](https:\u002F\u002Fdayihengliu.com)\u003C\u002Fli> \u003Cli>[Xingzhang Ren](https:\u002F\u002Fgithub.com\u002FXingzhangRen)\u003C\u002Fli> \u003Cli>[Bo Zheng](https:\u002F\u002Fgithub.com\u002FBakerBunker)\u003C\u002Fli> \u003Cli>[Rui Men](https:\u002F\u002Fgithub.com\u002FRuiMen)\u003C\u002Fli> \u003Cli>[Fan Zhou](https:\u002F\u002Fgithub.com\u002FFanZhou920)\u003C\u002Fli> \u003Cli>[Bowen Yu](https:\u002F\u002Fbowenyubig.github.io)\u003C\u002Fli> \u003Cli>[Jianxin Yang](https:\u002F\u002Fgithub.com\u002FJianxinYang)\u003C\u002Fli> \u003Cli>[Le Yu](https:\u002F\u002Fgithub.com\u002FYuLeAI)\u003C\u002Fli> \u003Cli>[Jingren Zhou](https:\u002F\u002Fwww.alibabagroup.com\u002Fen-US\u002Fnews\u002Fcto-jingren-zhou)\u003C\u002Fli> \u003Cli>[Junyang Lin](https:\u002F\u002Fgithub.com\u002Flinjunyang)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FQwenLM\u002FQwen3-Omni?style=social)](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen3-Omni) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.17765)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fqwen.ai\u002Fblog?id=65f766fc2dcba7905c1cb69cc4cab90e94126bf4&from=research.latest-advancements-list)\u003C\u002Fli>\u003Cli>[chat](https:\u002F\u002Fchat.qwen.ai\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FCV4E9rpNSD)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fqwenllm\u002Fqwen3-omni)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fhelp.aliyun.com\u002Fzh\u002Fmodel-studio\u002Fuser-guide\u002Fqwen-omni)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDao-AILab\u002Fflash-attention)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FQwen\u002FQwen3-Omni-Demo), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FQwen\u002FQwen3-Omni-Captioner-Demo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science-in-your-pocket\u002Fqwen3-omni-one-llm-for-text-images-audio-and-videos-aad51ea1a4e3)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FAlibaba_Qwen\u002Fstatus\u002F1970181599133344172)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_zdOrPju4_g), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0N8mif_OUlM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FCWSYLPJz8j0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FtRqEAN61qpk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FQwenLM\u002FQwen3-Omni\u002Fblob\u002Fmaster\u002Fcookbooks\u002Fimage_math.ipynb) | 22.09.2025 |\n| WMAR | Custom tokenizer-detokenizer finetuning procedure that improves RCC, and a complementary watermark synchronization layer | \u003Cul>\u003Cli>[Nikola Jovanović](https:\u002F\u002Fwww.sri.inf.ethz.ch\u002Fpeople\u002Fnikola)\u003C\u002Fli> \u003Cli>[Ismail Labiad](https:\u002F\u002Fgithub.com\u002Filabiad)\u003C\u002Fli> \u003Cli>[Tomáš Souček](https:\u002F\u002Fgithub.com\u002FsoCzech)\u003C\u002Fli> \u003Cli>[Martin Vechev](https:\u002F\u002Fwww.sri.inf.ethz.ch\u002Fpeople\u002Fmartin)\u003C\u002Fli> \u003Cli>[Pierre Fernandez](https:\u002F\u002Fpierrefdz.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fwmar?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fwmar) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.16349)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002FDiffPure), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fchameleon), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbytedance\u002F1d-tokenizer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmehrdadsaberi\u002Fwatermark_robustness)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fwmar\u002Fblob\u002Fmain\u002Fnotebooks\u002Fcolab.ipynb) | 19.09.2025 |\n| LIMIT | On the Theoretical Limitations of Embedding-Based Retrieval | \u003Cul>\u003Cli>[Orion Weller](https:\u002F\u002Forionweller.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michael Boratko](https:\u002F\u002Fwww.mboratko.com\u002F)\u003C\u002Fli> \u003Cli>[Iftekhar Naim](https:\u002F\u002Fgithub.com\u002Fiftekharnaim)\u003C\u002Fli> \u003Cli>[Jinhyuk Lee](https:\u002F\u002Fjhyuklee.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Flimit?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Flimit) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.21038)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fembeddings-benchmark\u002Fmteb)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Forionweller\u002FLIMIT), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Forionweller\u002FLIMIT-small)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fnoailabs.medium.com\u002Fon-the-theoretical-limitations-of-embedding-based-retrieval-new-research-paper-c70dc3edc817)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-deepmind\u002Flimit\u002Fblob\u002Fmaster\u002Fcode\u002Fgenerate_limit_dataset.ipynb) | 27.08.2025 |\n| DINOv3 | Produces high-quality dense features that achieve outstanding performance on various vision tasks, significantly surpassing previous self- and weakly-supervised foundation models | \u003Cul>\u003Cli>[Oriane Siméoni](https:\u002F\u002Forlanesimeoni.github.io\u002F)\u003C\u002Fli> \u003Cli>[Huy Vo](https:\u002F\u002Fhuyvvo.github.io\u002F)\u003C\u002Fli> \u003Cli>[Maximilian Seitzer](https:\u002F\u002Fgithub.com\u002FSeitzerM)\u003C\u002Fli> \u003Cli>[Federico Baldassarre](https:\u002F\u002Ffedebal.dev\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Maxime Oquab](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5vteYV8AAAAJ)\u003C\u002Fli> \u003Cli>[Cijo Jose](https:\u002F\u002Fgithub.com\u002Fcijojose)\u003C\u002Fli> \u003Cli>[Vasil Khalidov](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=tjazz3AAAAAJ)\u003C\u002Fli> \u003Cli>[Marc Szafraniec](https:\u002F\u002Fgithub.com\u002FMarcSzafraniec)\u003C\u002Fli> \u003Cli>[Seungeun Yi](https:\u002F\u002Fyi-seungeun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michaël Ramamonjisoa](https:\u002F\u002Fmichaelramamonjisoa.github.io\u002F)\u003C\u002Fli> \u003Cli>[Francisco Massa](https:\u002F\u002Fgithub.com\u002Ffmassa)\u003C\u002Fli> \u003Cli>[Daniel Haziza](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2eSKdFMAAAAJ)\u003C\u002Fli> \u003Cli>[Luca Wehrstedt](https:\u002F\u002Flucawehrstedt.info\u002F)\u003C\u002Fli> \u003Cli>[Jianyuan Wang](https:\u002F\u002Fjianyuanwang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Timothée Darcet](https:\u002F\u002Fgithub.com\u002FTimDarcet)\u003C\u002Fli> \u003Cli>[Théo Moutakanni](https:\u002F\u002Fgithub.com\u002FTheoMoutak)\u003C\u002Fli> \u003Cli>[Leonel Sentana](https:\u002F\u002Fleosentana.github.io\u002F)\u003C\u002Fli> \u003Cli>[Claire Roberts](https:\u002F\u002Fclaire-roberts.github.io\u002F)\u003C\u002Fli> \u003Cli>[Andrea Vedaldi](https:\u002F\u002Fandrea-vedaldi.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jamie Tolan](https:\u002F\u002Fjamietolan.github.io\u002F)\u003C\u002Fli> \u003Cli>[John Brandt](https:\u002F\u002Fjohnbrandt.github.io\u002F)\u003C\u002Fli> \u003Cli>[Camille Couprie](https:\u002F\u002Fcamillecouprie.github.io\u002F)\u003C\u002Fli> \u003Cli>[Julien Mairal](http:\u002F\u002Fthoth.inrialpes.fr\u002Fpeople\u002Fmairal\u002F)\u003C\u002Fli> \u003Cli>[Hervé Jégou](https:\u002F\u002Fgithub.com\u002Fjegou)\u003C\u002Fli> \u003Cli>[Patrick Labatut](https:\u002F\u002Fgithub.com\u002Fpatricklabatut)\u003C\u002Fli> \u003Cli>[Piotr Bojanowski](https:\u002F\u002Fgithub.com\u002Fpiotr-bojanowski)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fdinov3?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdinov3) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.10104), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.16334)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Ffacebook\u002Fdinov3-68924841bd6b561778e31009), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Fdinov3)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fdatasets\u002Fnikhil7280\u002Fcoco-image-caption)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science-in-your-pocket\u002Fmeta-dino-v3-the-ultimate-vision-ai-for-every-image-task-cf5ffc30a221)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fdinov3\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fdinov3-self-supervised-vision-model\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresources\u002Fmodels-and-libraries\u002Fdinov3-downloads\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-eOYWK6m3i8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fdinov3\u002Fblob\u002Fmain\u002Fnotebooks\u002Fsegmentation_tracking.ipynb) | 14.08.2025 |\n| Hogwild! Inference | Run LLM \"workers\" in parallel, allowing them to synchronize via a concurrently-updated attention cache and prompt these workers to decide how best to collaborate | \u003Cul>\u003Cli>[Gleb Rodionov](https:\u002F\u002Fgithub.com\u002Feqimp)\u003C\u002Fli> \u003Cli>[Roman Garipov](https:\u002F\u002Fgithub.com\u002Fgaripovroma)\u003C\u002Fli> \u003Cli>[Alina Shutova](https:\u002F\u002Fgithub.com\u002Fgoodevening13)\u003C\u002Fli> \u003Cli>[George Yakushev](https:\u002F\u002Fgithub.com\u002FMr-DarkTesla)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Vage Egiazarian](https:\u002F\u002Fgithub.com\u002FVahe1994)\u003C\u002Fli> \u003Cli>[Anton Sinitsin](https:\u002F\u002Fgithub.com\u002Fxtinkt)\u003C\u002Fli> \u003Cli>[Denis Kuznedelev](https:\u002F\u002Fgithub.com\u002FGodofnothing)\u003C\u002Fli> \u003Cli>[Dan Alistarh](https:\u002F\u002Fgithub.com\u002Fdalistarh)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Feqimp\u002Fhogwild_llm?style=social)](https:\u002F\u002Fgithub.com\u002Feqimp\u002Fhogwild_llm) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.06261), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.11903), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.04692), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.15337)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FQwen\u002FQwQ-32B)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fllama-4-multimodal-intelligence\u002F)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Feqimp.github.io\u002Fhogwild_llm\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLocalLLaMA\u002Fcomments\u002F1jv7x6l\u002Fhogwild_inference_parallel_llm_generation_via\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Feqimp\u002Fhogwild_llm\u002Fblob\u002Fmain\u002Fcolab_example.ipynb) | 15.07.2025 |\n| Grounding DINO | Marrying DINO with Grounded Pre-Training for Open-Set Object Detection | \u003Cul>\u003Cli>[Shilong Liu](https:\u002F\u002Fgithub.com\u002FSlongLiu)\u003C\u002Fli> \u003Cli>[Zhaoyang Zeng](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=U_cvvUwAAAAJ)\u003C\u002Fli> \u003Cli>[Tianhe Ren](https:\u002F\u002Frentainhe.github.io\u002F)\u003C\u002Fli> \u003Cli>[Feng Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ybRe9GcAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Hao Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=B8hPxMQAAAAJ)\u003C\u002Fli> \u003Cli>[Jie Yang](https:\u002F\u002Fyangjie-cv.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chunyuan Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Zd7WmXUAAAAJ)\u003C\u002Fli> \u003Cli>[Jianwei Yang](https:\u002F\u002Fjwyang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hang Su](https:\u002F\u002Fwww.suhangss.me\u002F)\u003C\u002Fli> \u003Cli>[Jun Zhu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=axsP38wAAAAJ)\u003C\u002Fli> \u003Cli>[Lei Zhang](https:\u002F\u002Fwww.leizhang.org\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FIDEA-Research\u002FGroundingDINO?style=social)](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002FGroundingDINO) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.05499)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002FDINO), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FUX-Decoder\u002FSemantic-SAM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOptimalScale\u002FDetGPT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002FOpenSeeD), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FUX-Decoder\u002FSegment-Everything-Everywhere-All-At-Once), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FX-Decoder\u002Ftree\u002Fxgpt), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002Fdetrex)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fzero-shot-object-detection-on-mscoco?p=grounding-dino-marrying-dino-with-grounded), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fzero-shot-object-detection-on-odinw?p=grounding-dino-marrying-dino-with-grounded), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fobject-detection-on-coco-minival?p=grounding-dino-marrying-dino-with-grounded), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fobject-detection-on-coco?p=grounding-dino-marrying-dino-with-grounded)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FwxWDt5UiwY8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FcMa77r3YrDk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FC4NqaRBz_Kw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FoEQYStnF2l8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Froboflow-ai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnotebooks\u002Fzero-shot-object-detection-with-grounding-dino.ipynb) | 10.07.2025 |\n| Hunyuan | Open-source large language model built on a fine-grained Mixture-of-Experts architecture | [manayang](https:\u002F\u002Fgithub.com\u002FManaEstras) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTencent-Hunyuan\u002FHunyuan-A13B?style=social)](https:\u002F\u002Fgithub.com\u002FTencent-Hunyuan\u002FHunyuan-A13B) \u003Cul>\u003Cli>[api](https:\u002F\u002Fcloud.tencent.com\u002Fproduct\u002Ftclm)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>]()\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FbsPcMEtV7v)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fhunyuaninfer\u002Fhunyuan-a13b\u002Ftags), [\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fhunyuaninfer\u002Fhunyuan-large\u002Ftags)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTencent\u002FAngelSlim)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ftencent\u002FHunyuan-A13B-Instruct)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLocalLLaMA\u002Fcomments\u002F1o22v1b\u002Fwhat_are_your_thoughts_on\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FTencentHunyuan\u002Fstatus\u002F1938525874904801490)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fhunyuan.tencent.com\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FH7JHky8Yh9Y), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYiQv6Av0jQg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQmRmcn_zOnU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBGtkuVKpVUA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdatawhalechina\u002Fself-llm\u002Fblob\u002Fmaster\u002Fmodels\u002FHunyuan-A13B-Instruct\u002F05-Hunyuan-A13B-Instruct-LoRA.ipynb) | 01.07.2025 |\n| Whisper | Automatic speech recognition system trained on 680,000 hours of multilingual and multitask supervised data collected from the web | \u003Cul>\u003Cli>[Alec Radford](http:\u002F\u002Fnewmu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jong Wook Kim](https:\u002F\u002Fjongwook.kim\u002F)\u003C\u002Fli> \u003Cli>[Tao Xu](https:\u002F\u002Fgithub.com\u002Fbayesian)\u003C\u002Fli> \u003Cli>[Greg Brockman](https:\u002F\u002Fgregbrockman.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Christine McLeavey](http:\u002F\u002Fchristinemcleavey.com\u002F)\u003C\u002Fli> \u003Cli>[Ilya Sutskever](http:\u002F\u002Fwww.cs.toronto.edu\u002F~ilya\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fopenai\u002Fwhisper?style=social)](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fwhisper) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.04356)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fopenai.com\u002Fresearch\u002Fwhisper)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkkroening\u002Fffmpeg-python)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FOCBZtgQGt1I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8SQV-B83tPU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FnE5iVtwKerA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fopenai\u002Fwhisper\u002Fblob\u002Fmaster\u002Fnotebooks\u002FLibriSpeech.ipynb) | 26.06.2025 |\n| IT³ | Idempotent Test-Time Training, approach that enables on-the-fly adaptation to distribution shifts using only the current test instance, without any auxiliary task design | \u003Cul>\u003Cli>[Nikita Durasov](https:\u002F\u002Fwww.norange.io\u002Fabout\u002F)\u003C\u002Fli> \u003Cli>[Assaf Shocher](https:\u002F\u002Fassafshocher.github.io\u002F)\u003C\u002Fli> \u003Cli>[Doruk Oner](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ESA2CsAAAAAJ)\u003C\u002Fli> \u003Cli>[Gal Chechik](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Fgal-chechik)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Alexei Efros](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~efros\u002F)\u003C\u002Fli> \u003Cli>[Pascal Fua](https:\u002F\u002Fpeople.epfl.ch\u002Fpascal.fua\u002Fbio)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fnikitadurasov\u002Fittt?style=social)](https:\u002F\u002Fgithub.com\u002Fnikitadurasov\u002Fittt) \u003Cul>\u003Cli>[ICML](https:\u002F\u002Ficml.cc\u002Fvirtual\u002F2025\u002Fposter\u002F45551)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.04201)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnikitadurasov\u002Ftorch-ttt)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.norange.io\u002Fprojects\u002Fittt\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Ftorch-ttt\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FeKGKpN8fFRM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fn5opIsl9WRA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fnikitadurasov\u002Fittt\u002Fblob\u002Fmain\u002Fexps\u002Fmnist\u002Fit3_torch_ttt.ipynb) | 25.06.2025 |\n| AlphaEvolve | Evolutionary coding agent that substantially enhances capabilities of state-of-the-art LLMs on highly challenging tasks such as tackling open scientific problems or optimizing critical pieces of computational infrastructure | \u003Cul>\u003Cli>[Alexander Novikov](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jMUkLqwAAAAJ)\u003C\u002Fli> \u003Cli>[Ngân Vu](https:\u002F\u002Flinkedin.com\u002Fin\u002Fthuynganvu)\u003C\u002Fli> \u003Cli>[Marvin Eisenberger](https:\u002F\u002Fcvg.cit.tum.de\u002Fmembers\u002Feisenber)\u003C\u002Fli> \u003Cli>[Emilien Dupont](https:\u002F\u002Femiliendupont.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Po-Sen Huang](https:\u002F\u002Fposenhuang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Adam Wagner](https:\u002F\u002Fzawagner22.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sergey Shirobokov](https:\u002F\u002Fgithub.com\u002Fshir994)\u003C\u002Fli> \u003Cli>[Borislav Kozlovskii](https:\u002F\u002Flinkedin.com\u002Fin\u002Fborislav-kozlovskii-63801b192)\u003C\u002Fli> \u003Cli>[Francisco Ruiz](https:\u002F\u002Ffranrruiz.github.io\u002F)\u003C\u002Fli> \u003Cli>[Abbas Mehrabian](https:\u002F\u002Fabbasmehrabian.com\u002F)\u003C\u002Fli> \u003Cli>[Pawan Kumar](https:\u002F\u002Fwww.ellis.ox.ac.uk\u002Fpeople\u002Fm-pawan-kumar)\u003C\u002Fli> \u003Cli>[Abigail See](https:\u002F\u002Fcs.stanford.edu\u002Fpeople\u002Fabisee\u002F)\u003C\u002Fli> \u003Cli>[Swarat Chaudhuri](https:\u002F\u002Fwww.cs.utexas.edu\u002F~swarat\u002F)\u003C\u002Fli> \u003Cli>[George Holland](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fg-aracil-holland)\u003C\u002Fli> \u003Cli>[Alex Davies](https:\u002F\u002Fwww.alexdavies.net\u002F)\u003C\u002Fli> \u003Cli>[Sebastian Nowozin](https:\u002F\u002Fwww.nowozin.net\u002Fsebastian\u002F)\u003C\u002Fli> \u003Cli>[Pushmeet Kohli](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FPushmeet_Kohli)\u003C\u002Fli> \u003Cli>[Matej Balog](http:\u002F\u002Fmatejbalog.eu\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Falphaevolve_results?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Falphaevolve_results) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fstorage.googleapis.com\u002Fdeepmind-media\u002FDeepMind.com\u002FBlog\u002Falphaevolve-a-gemini-powered-coding-agent-for-designing-advanced-algorithms\u002FAlphaEvolve.pdf), [\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fdiscover\u002Fblog\u002Falphaevolve-a-gemini-powered-coding-agent-for-designing-advanced-algorithms\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-alphaevolve-a-coding-agent-for-scientific-and-algorithmic-discovery-5732a876c2e2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fsingularity\u002Fcomments\u002F1kmhti8\u002Fdeepmind_introduces_alphaevolve_a_geminipowered\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FAlphaEvolve)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjCTvblRXnzg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FsGCmu7YKgPA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FT0eWBlFhFzc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDYyK76ZOUJU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FEMoiremdiA8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0-MA3jgYMMg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fx1FFLzTX-Kg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRH4hAgvYSzg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-deepmind\u002Falphaevolve_results\u002Fblob\u002Fmain\u002Fmathematical_results.ipynb) | 17.06.2025 |\n| V-JEPA 2 | Self-supervised approach that combines internet-scale video data with a small amount of interaction data, to develop models capable of understanding, predicting, and planning in the physical world | [FAIR](https:\u002F\u002Fai.meta.com\u002Fresearch\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fvjepa2?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fvjepa2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.09985)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fandlukyane.com\u002Fblog\u002Fpaper-review-vjepa2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Ffacebook\u002Fv-jepa-2-6841bad8413014e185b497a6)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-v-jepa-2-self-supervised-video-models-enable-understanding-prediction-and-planning-28410d8a1c6b)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fv-jepa-2-world-model-benchmarks)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FijEmo75zqZc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FsMO3BAPBLG4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fvjepa2\u002Fblob\u002Fmain\u002Fnotebooks\u002Fvjepa2_demo.ipynb) | 11.06.2025 |\n| TimesFM | Time-series foundation model for forecasting whose out-of-the-box zero-shot performance on a variety of public datasets comes close to the accuracy of state-of-the-art supervised forecasting models for each individual dataset | \u003Cul>\u003Cli>[Abhimanyu Das](https:\u002F\u002Fgithub.com\u002Fabhimanyudas747)\u003C\u002Fli> \u003Cli>[Weihao Kong](https:\u002F\u002Fweihaokong.github.io\u002F)\u003C\u002Fli> \u003Cli>[Rajat Sen](https:\u002F\u002Fgithub.com\u002Frajatsen91)\u003C\u002Fli> \u003Cli>[Yichen Zhou](https:\u002F\u002Fgithub.com\u002Fsiriuz42)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Ftimesfm?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Ftimesfm) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.10688)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fresearch.google\u002Fblog\u002Fa-decoder-only-foundation-model-for-time-series-forecasting\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Ftimesfm), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FSalesforce\u002FGIFT-Eval), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Fgoogle\u002Ftimesfm-release-66e4be5fdb56e960c1e482a6)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Ftimesfm\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_2SWu1SOcG0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYkzRP3xnMwc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F265Mpaj8O1U), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FOhEAS5oBcco)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FNielsRogge\u002FTransformers-Tutorials\u002Fblob\u002Fmaster\u002FTimesFM\u002FFine_tune_TimesFM_on_a_custom_dataset.ipynb) | 26.05.2025 |\n| Qwen2.5-Omni | End-to-end multimodal model designed to perceive diverse modalities, including text, images, audio, and video, while simultaneously generating text and natural speech responses in a streaming manner | \u003Cul>\u003Cli>[Jin Xu](https:\u002F\u002Fjxu-thu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhifang Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=LHzzlAgAAAAJ)\u003C\u002Fli> \u003Cli>[Jinzheng He](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=tsQVykcAAAAJ)\u003C\u002Fli> \u003Cli>[Ting He](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Yxat_NMAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Shuai Bai](https:\u002F\u002Fgithub.com\u002FShuaiBai623)\u003C\u002Fli> \u003Cli>[Keqin Chen](https:\u002F\u002Fgithub.com\u002FOliverChen0602)\u003C\u002Fli> \u003Cli>[Jialin Wang](https:\u002F\u002Fgithub.com\u002FJialinWangPKU)\u003C\u002Fli> \u003Cli>[Yang Fan](https:\u002F\u002Fgithub.com\u002Ffyabc)\u003C\u002Fli> \u003Cli>[Peng Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=7fjqA0YAAAAJ)\u003C\u002Fli> \u003Cli>[Bin Zhang](https:\u002F\u002Fzhangchbin.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xiong Wang](https:\u002F\u002Fgithub.com\u002Fwangxiongts)\u003C\u002Fli> \u003Cli>[Yunfei Chu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=41QhCyYAAAAJ)\u003C\u002Fli> \u003Cli>[Junyang Lin](https:\u002F\u002Fjustinlin610.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FQwenLM\u002FQwen2.5-Omni?style=social)](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen2.5-Omni) \u003Cul>\u003Cli>[API](https:\u002F\u002Fhelp.aliyun.com\u002Fzh\u002Fmodel-studio\u002Fuser-guide\u002Fqwen-omni)\u003C\u002Fli>\u003Cli>[Chat](https:\u002F\u002Fchat.qwen.ai\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.20215)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fqwenlm.github.io\u002Fblog\u002Fqwen2.5-omni\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FCV4E9rpNSD)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fqwenllm\u002Fqwen-omni)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDao-AILab\u002Fflash-attention)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002FQwen\u002Fqwen25-omni-67de1e5f0f9464dc6314b36e), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FQwen\u002FQwen2.5-Omni-7B-Demo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science-in-your-pocket\u002Fqwen2-5-omni-7b-input-anything-output-text-and-audio-llm-87c050716172)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyKcANdkRuNI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-JY1wcRRXMA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVez5TyC5YTE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5BqFvIm0joU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FQwenLM\u002FQwen2.5-Omni\u002Fblob\u002Fmain\u002Fcookbooks\u002Fomni_chatting_for_math.ipynb) | 29.04.2025 |\n| EAT | Emotional Adaptation for Audio-driven Talking-head method, which transforms emotion-agnostic talking-head models into emotion-controllable ones in a cost-effective and efficient manner through parameter-efficient adaptations | \u003Cul>\u003Cli>[Yuan Gan](https:\u002F\u002Fyuangan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zongxin Yang](https:\u002F\u002Fz-x-yang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xihang Yue](https:\u002F\u002Fgithub.com\u002Fyuexihang)\u003C\u002Fli> \u003Cli>[Lingyun Sun](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=zzW8d-wAAAAJ)\u003C\u002Fli> \u003Cli>[Yi Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RMSuNFwAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2fe6fd6c5a7e.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.02069) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyuangan\u002FEAT_code?style=social)](https:\u002F\u002Fgithub.com\u002Fyuangan\u002FEAT_code) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.04946)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyuangan\u002Fevaluation_eat), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYudongGuo\u002FAD-NeRF), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjixinya\u002FEAMM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhanglonghao1992\u002FOne-Shot_Free-View_Neural_Talking_Head_Synthesis), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFuxiVirtualHuman\u002FAAAI22-one-shot-talking-face), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHangz-nju-cuhk\u002FTalking-Face_PC-AVS), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fvid2vid)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyuangan.github.io\u002Feat\u002F)\u003C\u002Fli>\u003Cli>[video](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1sAoplzY4b6JCW0JQHf_HKEL5luuWuGAk\u002Fview)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Flp2nSLZp-88)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F133hwDHzsfRYl-nQCUQxJGjcXa5Fae22Z) | 22.04.2025 |\n| Moshi | Speech-text foundation model and full-duplex spoken dialogue framework | \u003Cul>\u003Cli>[Alexandre Défossez](https:\u002F\u002Fgithub.com\u002Fadefossez)\u003C\u002Fli> \u003Cli>[Laurent Mazaré](http:\u002F\u002Flaurentmazare.github.io)\u003C\u002Fli> \u003Cli>[Manu Orsini](https:\u002F\u002Fwww.facebook.com\u002Forsini9\u002F)\u003C\u002Fli> \u003Cli>[Amélie Royer](https:\u002F\u002Ffr.linkedin.com\u002Fin\u002Famélie-royer-aa26081a)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Patrick Pérez](https:\u002F\u002Fgithub.com\u002Fpppjer)\u003C\u002Fli> \u003Cli>[Hervé Jégou](https:\u002F\u002Fgithub.com\u002Fjerfju)\u003C\u002Fli> \u003Cli>[Edouard Grave](https:\u002F\u002Ffr.linkedin.com\u002Fin\u002Fedouard-grave-63099823)\u003C\u002Fli> \u003Cli>[Neil Zeghidour](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=p_aUHWgAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fkyutai-labs\u002Fmoshi?style=social)](https:\u002F\u002Fgithub.com\u002Fkyutai-labs\u002Fmoshi) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.00037), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.00037), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.03312), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2110.13900), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.14090)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fmoshi.chat\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkyutai-labs\u002Fmoshi-finetune), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZhangXInFD\u002FSpeechTokenizer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhaoheliu\u002FSemantiCodec-inference), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fencodec)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Fkyutai\u002Fmoshi-v01-release-66eaeaf3302bef6bd9ad7acd)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fmoshi\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1fmmm18\u002Fmoshi_a_speechtext_foundation_model_for_real_time\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0_c3bw_x6uU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FK1TbRsgwWd4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPO4EO7kUUQQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FedToLqBw_K8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDv31P1aTVOs)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fkyutai-labs\u002Fmoshi\u002Fblob\u002Fmaster\u002Fmoshi\u002Fdemo_moshi.ipynb) | 31.03.2025 |\n| BiRefNet | Bilateral reference framework for high-resolution dichotomous image segmentation | \u003Cul>\u003Cli>[Peng Zheng](https:\u002F\u002Fzhengpeng7.github.io\u002Fabout\u002F)\u003C\u002Fli> \u003Cli>[Dehong Gao](https:\u002F\u002Fteacher.nwpu.edu.cn\u002Fdehonggao)\u003C\u002Fli> \u003Cli>[Deng-Ping Fan](https:\u002F\u002Fdengpingfan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Li Liu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=9cMQrVsAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jorma Laaksonen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=qQP6WXIAAAAJ)\u003C\u002Fli> \u003Cli>[Wanli Ouyang](https:\u002F\u002Fwlouyang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nicu Sebe](https:\u002F\u002Fdisi.unitn.it\u002F~sebe\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_b8af1a40a030.png)](https:\u002F\u002Fdoi.org\u002F10.26599\u002FAIR.2024.9150038) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FZhengPeng7\u002FBiRefNet?style=social)](https:\u002F\u002Fgithub.com\u002FZhengPeng7\u002FBiRefNet) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.03407), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2302.14485)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002Fd9NN5sgFrq)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FKazuhito00\u002FBiRefNet-ONNX-Sample), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZHO-ZHO-ZHO\u002FComfyUI-BiRefNet-ZHO), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fviperyl\u002FComfyUI-BiRefNet)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FZhengPeng7\u002FBiRefNet_demo), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FZhengPeng7\u002FBiRefNet)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.birefnet.top\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fdichotomous-image-segmentation-on-dis-te1?p=bilateral-reference-for-high-resolution), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fcamouflaged-object-segmentation-on-cod?p=bilateral-reference-for-high-resolution), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Frgb-salient-object-detection-on-davis-s?p=bilateral-reference-for-high-resolution)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1B6aKZ3ekcvKMkSBn0N5mCASLUYMp0whK) | 24.03.2025 |\n| ESM | Evolutionary Scale Modeling: Pretrained language models for proteins | \u003Cul>\u003Cli>[Zeming Lin](https:\u002F\u002Fresearch.facebook.com\u002Fpeople\u002Flin-zeming\u002F)\u003C\u002Fli> \u003Cli>[Roshan Rao](https:\u002F\u002Frmrao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Brian Hie](https:\u002F\u002Fbrianhie.com\u002F)\u003C\u002Fli> \u003Cli>[Zhongkai Zhu](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fzhongkai-zhu-03a27424)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Allan dos Santos Costa](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Zb4RsFsAAAAJ)\u003C\u002Fli> \u003Cli>[Maryam Fazel-Zarandi](https:\u002F\u002Fwww.maryamfazel.com\u002F)\u003C\u002Fli> \u003Cli>[Tom Sercu](https:\u002F\u002Ftom.sercu.me\u002F)\u003C\u002Fli> \u003Cli>[Salvatore Candido](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=BDgbhmEAAAAJ)\u003C\u002Fli> \u003Cli>[Alexander Rives](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vqb78-gAAAAJ)\u003C\u002Fli> \u003Cli>[Joshua Meier](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2M0OltAAAAAJ)\u003C\u002Fli> \u003Cli>[Robert Verkuil](https:\u002F\u002Fdblp.org\u002Fpid\u002F296\u002F8930.html)\u003C\u002Fli> \u003Cli>[Jason Liu](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fliujiayi\u002F)\u003C\u002Fli> \u003Cli>[Chloe Hsu](https:\u002F\u002Fchloe-hsu.com\u002F)\u003C\u002Fli> \u003Cli>[Adam Lerer](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Ad6O4-0AAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9220bdc4ae8c.png)](https:\u002F\u002Fdoi.org\u002F10.1101\u002F622803) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fesm?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fesm) \u003Cul>\u003Cli>[ESM Atlas](https:\u002F\u002Fesmatlas.com\u002F)\u003C\u002Fli>\u003Cli>[FSDP](https:\u002F\u002Ffairscale.readthedocs.io\u002Fen\u002Fstable\u002Fapi\u002Fnn\u002Ffsdp.html)\u003C\u002Fli>\u003Cli>[ICML](https:\u002F\u002Fproceedings.mlr.press\u002Fv139\u002Frao21a.html)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fftp.uniprot.org\u002Fpub\u002Fdatabases\u002Funiprot\u002Fprevious_releases\u002Frelease-2018_03\u002Funiref\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsokrypton\u002FColabFold)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Fesm)\u003C\u002Fli>\u003Cli>[paper](https:\u002F\u002Fdoi.org\u002F10.1101\u002F2022.07.20.500902), [paper](https:\u002F\u002Fdoi.org\u002F10.1101\u002F2021.07.09.450648), [paper](https:\u002F\u002Fdoi.org\u002F10.1101\u002F2022.04.10.487779), [paper](https:\u002F\u002Fdoi.org\u002F10.1101\u002F2022.12.21.521521)\u003C\u002Fli>\u003Cli>[pubmed](https:\u002F\u002Fpubmed.ncbi.nlm.nih.gov\u002F33876751\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FN-eisTvUYrk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGHoE4VkDehY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsokrypton\u002FColabFold\u002Fblob\u002Fmain\u002FESMFold.ipynb) | 21.03.2025 |\n| Video Seal | Comprehensive framework for neural video watermarking and a competitive open-sourced model | \u003Cul>\u003Cli>[Pierre Fernandez](https:\u002F\u002Fpierrefdz.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hady Elsahar](https:\u002F\u002Fwww.hadyelsahar.io\u002F)\u003C\u002Fli> \u003Cli>[Zeki Yalniz](https:\u002F\u002Fai.meta.com\u002Fpeople\u002F274060152308398\u002Fi-zeki-yalniz\u002F)\u003C\u002Fli> \u003Cli>[Alexandre Mourachko](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=OD9-erYAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fvideoseal?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fvideoseal) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.09492), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.20468)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Faidemos.meta.com\u002Fvideoseal)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresearch\u002Fpublications\u002Fvideo-seal-open-and-efficient-video-watermarking\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fvideoseal\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fvideoseal\u002Fblob\u002Fmain\u002Fnotebooks\u002Fcolab.ipynb) | 17.03.2025 |\n| SigLIP 2 | Family of new multilingual vision-language encoders that build on the success of the original SigLIP | \u003Cul>\u003Cli>[Michael Tschannen](https:\u002F\u002Fmitscha.github.io\u002F)\u003C\u002Fli> \u003Cli>[Alexey Gritsenko](https:\u002F\u002Fgithub.com\u002FAlexeyG)\u003C\u002Fli> \u003Cli>[Xiao Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ukyXqzMAAAAJ)\u003C\u002Fli> \u003Cli>[Muhammad Ferjad Naeem](https:\u002F\u002Fferjad.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ibrahim Alabdulmohsin](https:\u002F\u002Fibomohsin.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nikhil Parthasarathy](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=X9mO4ckAAAAJ)\u003C\u002Fli> \u003Cli>[Talfan Evans](https:\u002F\u002Ftalfanevans.co.uk\u002F)\u003C\u002Fli> \u003Cli>[Lucas Beyer](http:\u002F\u002Flucasb.eyer.be)\u003C\u002Fli> \u003Cli>[Ye Xia](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QQhJ1pAAAAAJ)\u003C\u002Fli> \u003Cli>[Basil Mustafa](https:\u002F\u002Fwww.basilmustafa.com\u002F)\u003C\u002Fli> \u003Cli>[Olivier Hénaff](https:\u002F\u002Fwww.olivierhenaff.com\u002F)\u003C\u002Fli> \u003Cli>[Jeremiah Harmsen](https:\u002F\u002Fresearch.google\u002Fpeople\u002Fjeremiahharmsen)\u003C\u002Fli> \u003Cli>[Andreas Steiner](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vIZeAu4AAAAJ)\u003C\u002Fli> \u003Cli>[Xiaohua Zhai](https:\u002F\u002Fsites.google.com\u002Fview\u002Fxzhai)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fbig_vision?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fbig_vision\u002Fblob\u002Fmain\u002Fbig_vision\u002Fmodels\u002Fvit.py) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.14786), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.15343)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fbig_vision\u002Fblob\u002Fmain\u002Fbig_vision\u002Fmodels\u002Fproj\u002Fimage_text\u002Ftwo_towers.py), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fbig_vision\u002Fblob\u002Fmain\u002Fbig_vision\u002Fmodels\u002Fproj\u002Fimage_text\u002Fnaflex_vit.py), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fbig_vision\u002Fblob\u002Fmain\u002Fbig_vision\u002Fpp\u002Fproj\u002Fimage_text\u002Fops_naflex.py)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Fgoogle\u002Fsiglip2-67b5dcef38c175486e240107)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-siglip-2-multilingual-vision-language-encoders-with-improved-semantic-understanding-b7b578002adc), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fritvik19.medium.com\u002Fpapers-explained-320-siglip-2-dba08ff09559)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fr8y7gXIpb4A)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Fbig_vision\u002Fblob\u002Fmain\u002Fbig_vision\u002Fconfigs\u002Fproj\u002Fimage_text\u002FSigLIP2_demo.ipynb) | 17.03.2025 |\n| DeepLabCut | Efficient method for markerless pose estimation based on transfer learning with deep neural networks that achieves excellent results with minimal training data | \u003Cul>\u003Cli>[Alexander Mathis](https:\u002F\u002Fgithub.com\u002FAlexEMG)\u003C\u002Fli> \u003Cli>[Pranav Mamidanna](https:\u002F\u002Fpranavm19.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kevin Cury](https:\u002F\u002Fkevincury.com\u002F)\u003C\u002Fli> \u003Cli>[Taiga Abe](https:\u002F\u002Fcellistigs.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Venkatesh Murthy](https:\u002F\u002Fgithub.com\u002Fvenkateshnmurthy)\u003C\u002Fli> \u003Cli>[Mackenzie Mathis](https:\u002F\u002Fgithub.com\u002FMMathisLab)\u003C\u002Fli> \u003Cli>[Matthias Bethge](https:\u002F\u002Fbethgelab.org\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_e9c6ba8d4911.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41593-018-0209-y) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FDeepLabCut\u002FDeepLabCut?style=social)](https:\u002F\u002Fgithub.com\u002FDeepLabCut\u002FDeepLabCut) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1605.03170), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.03142), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.11229), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2009.00564), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.13868), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.13868)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fdeeplabcut\u002Fdeeplabcut)\u003C\u002Fli>\u003Cli>[forum](https:\u002F\u002Fforum.image.sc\u002Ftag\u002Fdeeplabcut)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDeepLabCut\u002FDLCutils), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDeepLabCut\u002FDeepLabCut-Workshop-Materials)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@cziscience\u002Fhow-open-source-software-contributors-are-accelerating-biomedicine-1a5f50f6846a)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fdeeplabcut\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002FDeepLabCut)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fwww.deeplabcut.org\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002F@deeplabcut7702), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuWZu3rnj-kQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTeb5r2TNAYs)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FDeepLabCut\u002FDeepLabCut\u002Fblob\u002Fmaster\u002Fexamples\u002FCOLAB\u002FCOLAB_maDLC_TrainNetwork_VideoAnalysis.ipynb) | 28.02.2025 |\n| STAR | Spatial Temporal Augmentation with T2V models for Real-world video super-resolution, a novel approach that leverages T2V models for real-world video super-resolution, achieving realistic spatial details and robust temporal consistency | \u003Cul>\u003Cli>[Rui Xie](https:\u002F\u002Fgithub.com\u002FCSRuiXie)\u003C\u002Fli> \u003Cli>[Yinhong Liu](https:\u002F\u002Fgithub.com\u002Fyhliu04)\u003C\u002Fli> \u003Cli>[Penghao Zhou](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=yWq1Fd4AAAAJ)\u003C\u002Fli> \u003Cli>[Chen Zhao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Uhp3JKgAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jun Zhou](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=w03CHFwAAAAJ)\u003C\u002Fli> \u003Cli>[Kai Zhang](https:\u002F\u002Fcszn.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhenyu Zhang](https:\u002F\u002Fjessezhang92.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jian Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=6CIDtZQAAAAJ)\u003C\u002Fli> \u003Cli>[Zhenheng Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Ds5wwRoAAAAJ)\u003C\u002Fli> \u003Cli>[Ying Tai](https:\u002F\u002Ftyshiwo.github.io\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNJU-PCALab\u002FSTAR?style=social)](https:\u002F\u002Fgithub.com\u002FNJU-PCALab\u002FSTAR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.02976)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fali-vilab\u002FVGen), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FVchitect\u002FVEnhancer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNJU-PCALab\u002FOpenVid-1M), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhpcaitech\u002FOpen-Sora\u002Ftree\u002Fmain\u002Ftools\u002Fcaption#pllava-captioning)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FSherryX\u002FSTAR)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-star-spatial-temporal-augmentation-with-text-to-video-models-for-real-world-video-ff0ddcc6352f)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnju-pcalab.github.io\u002Fprojects\u002FSTAR)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fhx0zrql-SrU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1K8A1U_BNpAteRhhW9A8pAYs6LWjItQs_) | 22.01.2025 |\n| InvSR | Image super-resolution technique based on diffusion inversion, aiming at harnessing the rich image priors encapsulated in large pre-trained diffusion models to improve SR performance | \u003Cul>\u003Cli>[Zongsheng Yue](https:\u002F\u002Fzsyoaoa.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kang Liao](https:\u002F\u002Fkangliao929.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FzsyOAOA\u002FInvSR?style=social)](https:\u002F\u002Fgithub.com\u002FzsyOAOA\u002FInvSR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.09013)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcsjcai\u002FRealSR)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FOAOA\u002FInvSR), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fstabilityai\u002Fsd-turbo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FmCd7ODSmols)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1hjgCFnAU4oUUhh9VRfTwsFN1AiIjdcSR) | 21.01.2025 |\n| ModernBERT | Bringing modern model optimizations to encoder-only models and representing a major Pareto improvement over older encoders | \u003Cul>\u003Cli>[Benjamin Warner](https:\u002F\u002Fbenjaminwarner.dev\u002F)\u003C\u002Fli> \u003Cli>[Antoine Chaffin](https:\u002F\u002Fantoine.chaffin.fr\u002F)\u003C\u002Fli> \u003Cli>[Benjamin Clavié](https:\u002F\u002Fben.clavie.eu\u002F)\u003C\u002Fli> \u003Cli>[Orion Weller](https:\u002F\u002Forionweller.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Oskar Hallström](https:\u002F\u002Fgithub.com\u002Fohallstrom)\u003C\u002Fli> \u003Cli>[Said Taghadouini](https:\u002F\u002Fgithub.com\u002Fstaghado)\u003C\u002Fli> \u003Cli>[Alexis Gallagher](https:\u002F\u002Falexisgallagher.com\u002F)\u003C\u002Fli> \u003Cli>[Raja Biswas](https:\u002F\u002Fgithub.com\u002Frbiswasfc)\u003C\u002Fli> \u003Cli>[Faisal Ladhak](https:\u002F\u002Fgithub.com\u002Ffladhak)\u003C\u002Fli> \u003Cli>[Tom Aarsen](https:\u002F\u002Fwww.tomaarsen.com\u002Fhome)\u003C\u002Fli> \u003Cli>[Nathan Cooper](https:\u002F\u002Fnathancooper.io\u002F)\u003C\u002Fli> \u003Cli>[Griffin Adams](https:\u002F\u002Fgithub.com\u002Fgriff4692)\u003C\u002Fli> \u003Cli>[Jeremy Howard](https:\u002F\u002Fjeremy.fast.ai\u002F)\u003C\u002Fli> \u003Cli>[Iacopo Poli](https:\u002F\u002Fgithub.com\u002Fiacolippo)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAnswerDotAI\u002FModernBERT?style=social)](https:\u002F\u002Fgithub.com\u002FAnswerDotAI\u002FModernBERT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.13663)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmosaicml\u002Fexamples\u002Ftree\u002Fmain\u002Fexamples\u002Fbenchmarks\u002Fbert), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flightonai\u002Fpylate)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fmodernbert)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@sahin.samia\u002Fmodernbert-the-next-generation-of-encoder-models-a-guide-to-using-and-fine-tuning-for-nlp-tasks-995a50d5232f)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FZ1Dl3juwtSU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-kXUWeNcfUw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FZWo6Q8580sA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F92HKsDHD9XI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAnswerDotAI\u002FModernBERT\u002Fblob\u002Fmaster\u002Fexamples\u002Ffinetune_modernbert_on_glue.ipynb) | 22.12.2024 |\n| GraphCast | Learning skillful medium-range global weather forecasting | \u003Cul>\u003Cli>[Rémi Lam](https:\u002F\u002Fgithub.com\u002Fremilam)\u003C\u002Fli> \u003Cli>[Alvaro Sanchez-Gonzalez](https:\u002F\u002Fgithub.com\u002Falvarosg)\u003C\u002Fli> \u003Cli>[Matthew Willson](https:\u002F\u002Fgithub.com\u002Fmjwillson)\u003C\u002Fli> \u003Cli>[Peter Wirnsberger](https:\u002F\u002Fpewi.org\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Meire Fortunato](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=_fMHSIUAAAAJ)\u003C\u002Fli> \u003Cli>[Ferran Alet](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=1lmBq3QAAAAJ)\u003C\u002Fli> \u003Cli>[Suman Ravuri](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsuman-ravuri-81928082)\u003C\u002Fli> \u003Cli>[Timo Ewalds](https:\u002F\u002Fgithub.com\u002Ftewalds)\u003C\u002Fli> \u003Cli>[Zach Eaton-Rosen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=mQ3zD_wAAAAJ)\u003C\u002Fli> \u003Cli>[Weihua Hu](https:\u002F\u002Fweihua916.github.io\u002F)\u003C\u002Fli> \u003Cli>[Alexander Merose](https:\u002F\u002Falex.merose.com\u002F)\u003C\u002Fli> \u003Cli>[Stephan Hoyer](https:\u002F\u002Fstephanhoyer.com\u002F)\u003C\u002Fli> \u003Cli>[George Holland](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fg-aracil-holland)\u003C\u002Fli> \u003Cli>[Oriol Vinyals](https:\u002F\u002Fresearch.google\u002Fpeople\u002Foriol-vinyals\u002F)\u003C\u002Fli> \u003Cli>[Jacklynn Stott](https:\u002F\u002Flinkedin.com\u002Fin\u002Fjacklynnstott)\u003C\u002Fli> \u003Cli>[Alexander Pritzel](https:\u002F\u002Fgithub.com\u002Fa-pritzel)\u003C\u002Fli> \u003Cli>[Shakir Mohamed](https:\u002F\u002Fwww.shakirm.com\u002F)\u003C\u002Fli> \u003Cli>[Peter Battaglia](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=nQ7Ij30AAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_95a3b8c45abb.png)](https:\u002F\u002Fdoi.org\u002F10.1126\u002Fscience.adi2336) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Fgraphcast?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fgraphcast) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.12794)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fwww.ecmwf.int\u002Fen\u002Fforecasts\u002Fdatasets\u002Freanalysis-datasets\u002Fera5)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fdiscover\u002Fblog\u002Fgraphcast-ai-model-for-faster-and-more-accurate-global-weather-forecasting\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fchex), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdask\u002Fdask), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fjaxline), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Ftree), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmikedh\u002Ftrimesh)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Ftowardsdatascience.com\u002Fgraphcast-how-to-get-things-done-f2fd5630c5fb)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBufUW7h9TB8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPD1v5PCJs_o), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FEul-JN9Nwb0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBTyhgp9Hugc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FaJ_H4exg0xU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Fgraphcast\u002Fblob\u002Fmaster\u002Fgraphcast_demo.ipynb) | 04.12.2024 |\n| TAPIR | Tracking Any Point with per-frame Initialization and temporal Refinement | \u003Cul>\u003Cli>[Carl Doersch](http:\u002F\u002Fwww.carldoersch.com\u002F)\u003C\u002Fli> \u003Cli>[Yi Yang](https:\u002F\u002Fyangyi02.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mel Vecerik](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Jvi_XPAAAAAJ)\u003C\u002Fli> \u003Cli>[Dilara Gokay](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=cnbENAEAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ankush Gupta](https:\u002F\u002Fankushgupta.org\u002F)\u003C\u002Fli> \u003Cli>[Yusuf Aytar](https:\u002F\u002Fpeople.csail.mit.edu\u002Fyusuf\u002F)\u003C\u002Fli> \u003Cli>[Joao Carreira](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=IUZ-7_cAAAAJ)\u003C\u002Fli> \u003Cli>[Andrew Zisserman](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~az\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Ftapnet?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Ftapnet) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.08637), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.15975)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fdeepmind-tapir.github.io\u002F), [blog post](https:\u002F\u002Fdeepmind-tapir.github.io\u002Fblogpost.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fwww.deepmind.com\u002Fopen-source\u002Fkinetics)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fkubric\u002Ftree\u002Fmain\u002Fchallenges\u002Fpoint_tracking)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@jumabek4044\u002Fwhat-is-tapir-tracking-any-point-with-per-frame-initialization-and-temporal-refinement-and-how-it-bdad9946dc53)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002F58168e8a92994655d6da3939e7cc0918-Abstract-Datasets_and_Benchmarks.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2HSHofqoJ9M), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FI1DQJH3v7Nk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Ftapnet\u002Fblob\u002Fmaster\u002Fcolabs\u002Fcausal_tapir_demo.ipynb) | 30.11.2024 |\n| ConsisID | Tuning-free DiT-based controllable IPT2V model to keep human identity consistent in the generated video | \u003Cul>\u003Cli>[Shenghai Yuan](https:\u002F\u002Fshyuanbest.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jinfa Huang](https:\u002F\u002Finfaaa.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xianyi He](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=f72qwN8AAAAJ)\u003C\u002Fli> \u003Cli>[Yunyuan Ge](https:\u002F\u002Fgithub.com\u002Fyunyangge)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yujun Shi](https:\u002F\u002Fyujun-shi.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liuhan Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=eALObLQAAAAJ)\u003C\u002Fli> \u003Cli>[Jiebo Luo](https:\u002F\u002Fwww.cs.rochester.edu\u002Fu\u002Fjluo\u002F)\u003C\u002Fli> \u003Cli>[Li Yuan](https:\u002F\u002Fyuanli2333.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPKU-YuanGroup\u002FConsisID?style=social)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FConsisID) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.17440)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FXueZeyue\u002FDanceGRPO), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpenS2V-Nexus), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fali-vilab\u002FTeaCache), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxdit-project\u002FxDiT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkijai\u002FComfyUI-CogVideoXWrapper), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffeizc\u002FIngredients), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Faigc-apps\u002FEasyAnimate), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Faigc-apps\u002FVideoX-Fun), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjunjiehe96\u002FUniPortrait)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FBestWishYsh\u002FConsisID-preview-Data), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FBestWishYsh\u002FConsisID-preview-Space)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpku-yuangroup.github.io\u002FConsisID\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FAdinaYakup\u002Fstatus\u002F1862604191631573122), [\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002Fcamenduru\u002Fstatus\u002F1861957812152078701)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPhlgC-bI5SQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FConsisID-jupyter\u002Fblob\u002Fmain\u002FConsisID_jupyter.ipynb) | 28.11.2024 |\n| T2M-GPT | Conditional generative framework based on Vector Quantised-Variational AutoEncoder and Generative Pre-trained Transformer for human motion generation from textural descriptions | \u003Cul>\u003Cli>[Jianrong Zhang](https:\u002F\u002Fgithub.com\u002FJiro-zhang)\u003C\u002Fli> \u003Cli>[Yangsong Zhang](https:\u002F\u002Fgithub.com\u002FMael-zys)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002Facademic\u002F)\u003C\u002Fli> \u003Cli>[Shaoli Huang](https:\u002F\u002Fshaoli-huang.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yong Zhang](https:\u002F\u002Fyzhang2016.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hongwei Zhao](https:\u002F\u002Fteachers.jlu.edu.cn\u002Fzhaohongwei\u002Fen\u002Findex.htm)\u003C\u002Fli> \u003Cli>[Hongtao Lu](https:\u002F\u002Fwww.cs.sjtu.edu.cn\u002Fen\u002FPeopleDetail.aspx?id=156)\u003C\u002Fli> \u003Cli>[Xi Shen](https:\u002F\u002Fxishen0220.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c2afe1505606.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52729.2023.01415) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FMael-zys\u002FT2M-GPT?style=social)](https:\u002F\u002Fgithub.com\u002FMael-zys\u002FT2M-GPT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.06052)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEricGuo5513\u002FHumanML3D), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEricGuo5513\u002Ftext-to-motion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FGuyTevet\u002Fmotion-diffusion-model), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEricGuo5513\u002FTM2T)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fvumichien\u002FT2M-GPT), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fvumichien\u002Fgenerate_human_motion)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@kaveh.kamali\u002Ft2m-gpt-pioneering-human-motion-generation-from-textual-descriptions-48dc62b5cd7a)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmael-zys.github.io\u002FT2M-GPT\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F09K2cx9P0_0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Vy69w2q2d-Hg19F-KibqG0FRdpSj3L4O) | 24.11.2024 |\n| PuLID | Pure and Lightning ID customization, a tuning-free ID customization method for text-to-image generation | \u003Cul>\u003Cli>[Zinan Guo](https:\u002F\u002Fgithub.com\u002Fguozinan126)\u003C\u002Fli> \u003Cli>[Yanze Wu](https:\u002F\u002Ftothebeginning.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhuowei Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ow1jGJkAAAAJ)\u003C\u002Fli> \u003Cli>[Lang Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=h5xex20AAAAJ)\u003C\u002Fli> \u003Cli>[Qian He](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=9rWWCgUAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FToTheBeginning\u002FPuLID?style=social)](https:\u002F\u002Fgithub.com\u002FToTheBeginning\u002FPuLID) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.16022)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcubiq\u002FPuLID_ComfyUI), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZHO-ZHO-ZHO\u002FComfyUI-PuLID-ZHO), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMikubill\u002Fsd-webui-controlnet\u002Fpull\u002F2838)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fcomfyui\u002Fcomments\u002F1cnv269\u002Fpulid_pure_and_lightning_id_customization_via\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FPuLID-jupyter\u002Fblob\u002Fmain\u002FPuLID_jupyter.ipynb) | 09.11.2024 |\n| CoTracker | Architecture that jointly tracks multiple points throughout an entire video | \u003Cul>\u003Cli>[Nikita Karaev](https:\u002F\u002Fnikitakaraevv.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ignacio Rocco](https:\u002F\u002Fwww.irocco.info\u002F)\u003C\u002Fli> \u003Cli>[Benjamin Graham](https:\u002F\u002Fai.meta.com\u002Fpeople\u002Fbenjamin-graham\u002F)\u003C\u002Fli> \u003Cli>[Natalia Neverova](https:\u002F\u002Fnneverova.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Andrea Vedaldi](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vedaldi\u002F)\u003C\u002Fli> \u003Cli>[Christian Rupprecht](https:\u002F\u002Fchrirupp.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fco-tracker?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fco-tracker) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.07635), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.11898)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbenjiebob\u002FBADJA)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fco-tracker.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fw5QVc7BVGPA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fco-tracker\u002Fblob\u002Fmain\u002Fnotebooks\u002Fdemo.ipynb) | 16.10.2024 |\n| PIFu | Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization | \u003Cul>\u003Cli>[Ryota Natsume](https:\u002F\u002Fgithub.com\u002Fnanopoteto)\u003C\u002Fli> \u003Cli>[Shunsuke Saito](https:\u002F\u002Fshunsukesaito.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zeng Huang](https:\u002F\u002Fzeng.science\u002F)\u003C\u002Fli> \u003Cli>[Angjoo Kanazawa](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~kanazawa\u002F)\u003C\u002Fli> \u003Cli>[Hao Li](http:\u002F\u002Fhao.li)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_d66c3b54eb70.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV.2019.00239) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fshunsukesaito\u002FPIFu?style=social)](https:\u002F\u002Fgithub.com\u002Fshunsukesaito\u002FPIFu) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.05172)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=S1FpjwKqtPs)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1GFSsqP2BWz4gtq0e-nki00ZHSirXwFyY) | 08.10.2024 |\n| DifFace | Method that is capable of coping with unseen and complex degradations more gracefully without complicated loss designs | \u003Cul>\u003Cli>[Zongsheng Yue](https:\u002F\u002Fzsyoaoa.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_28a1a4fb4867.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTPAMI.2024.3432651) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FzsyOAOA\u002FDifFace?style=social)](https:\u002F\u002Fgithub.com\u002FzsyOAOA\u002FDifFace) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.06512)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fimproved-diffusion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepcam-cn\u002Fyolov5-face), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002Ffacexlib)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FOAOA\u002FDifFace)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1BNtoPPRuJwNDvqfwDOOmD9XJyF05Zh4m) | 05.10.2024 |\n| Segment Anything 2 | Foundation model towards solving promptable visual segmentation in images and videos | \u003Cul>\u003Cli>[Nikhila Ravi](https:\u002F\u002Fnikhilaravi.com\u002F)\u003C\u002Fli> \u003Cli>[Valentin Gabeur](https:\u002F\u002Fgabeur.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuan-Ting Hu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=E8DVVYQAAAAJ)\u003C\u002Fli> \u003Cli>[Ronghang Hu](https:\u002F\u002Fronghanghu.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chaitanya Ryali](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4LWx24UAAAAJ)\u003C\u002Fli> \u003Cli>[Tengyu Ma](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=VeTSl0wAAAAJ)\u003C\u002Fli> \u003Cli>[Haitham Khedr](https:\u002F\u002Fhkhedr.com\u002F)\u003C\u002Fli> \u003Cli>[Roman Rädle](https:\u002F\u002Fscholar.google.de\u002Fcitations?user=Tpt57v0AAAAJ)\u003C\u002Fli> \u003Cli>[Chloé Rolland](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=n-SnMhoAAAAJ)\u003C\u002Fli> \u003Cli>[Laura Gustafson](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=c8IpF9gAAAAJ)\u003C\u002Fli> \u003Cli>[Eric Mintun](https:\u002F\u002Fericmintun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Junting Pan](https:\u002F\u002Fjunting.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kalyan Vasudev](lwala](https:\u002F\u002Fscholar.google.co.in\u002Fcitations?user=m34oaWEAAAAJ)\u003C\u002Fli> \u003Cli>[Nicolas Carion](https:\u002F\u002Fwww.nicolascarion.com\u002F)\u003C\u002Fli> \u003Cli>[Chao-Yuan](u](https:\u002F\u002Fchaoyuan.org\u002F)\u003C\u002Fli> \u003Cli>[Ross Girshick](https:\u002F\u002Fwww.rossgirshick.info\u002F)\u003C\u002Fli> \u003Cli>[Piotr Dollár](https:\u002F\u002Fpdollar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Christoph Feichtenhofer](https:\u002F\u002Ffeichtenhofer.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fsegment-anything-2?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fsegment-anything-2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.00714)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fsam2.metademolab.com\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzsef123\u002FConnected_components_PyTorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fmodels?search=facebook\u002Fsam2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresearch\u002Fpublications\u002Fsam-2-segment-anything-in-images-and-videos\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fdatasets\u002Fsegment-anything-video), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fsegment-anything-2)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fai.meta.com\u002Fsam2\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FAIatMeta\u002Fstatus\u002F1818055906179105010)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=w-cmMcMZoZ4&t=2325s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FO8QdvZbRDp4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FDv003fTyO-Y), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIW7jFq3vQbw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fsegment-anything-2\u002Fblob\u002Fmain\u002Fnotebooks\u002Fimage_predictor_example.ipynb) | 01.10.2024 |\n| Open-Unmix | A deep neural network reference implementation for music source separation, applicable for researchers, audio engineers and artists | \u003Cul>\u003Cli>[Fabian-Robert Stöter](http:\u002F\u002Ffaroit.com\u002F)\u003C\u002Fli> \u003Cli>[Antoine Liutkus](https:\u002F\u002Fgithub.com\u002Faliutkus)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_7aec23c2275d.png)](https:\u002F\u002Fdoi.org\u002F10.21105\u002Fjoss.01667) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsigsep\u002Fopen-unmix-pytorch?style=social)](https:\u002F\u002Fgithub.com\u002Fsigsep\u002Fopen-unmix-pytorch) \u003Cul>\u003Cli>[data](https:\u002F\u002Fsigsep.github.io\u002Fdatasets\u002Fmusdb.html#musdb18-compressed-stems)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsigsep\u002Fnorbert)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsigsep.github.io\u002Fopen-unmix\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fmusic-source-separation-on-musdb18?p=open-unmix-a-reference-implementation-for)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLhA3b2k8R3t0VpYCpCTU2B1h604rvnV4N)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1mijF0zGWxN-KaxTnd0q6hayAlrID5fEQ) | 25.09.2024 |\n| Deep Painterly Harmonization | Algorithm produces significantly better results than photo compositing or global stylization techniques and that it enables creative painterly edits that would be otherwise difficult to achieve | \u003Cul>\u003Cli>[Fujun Luan](https:\u002F\u002Fluanfujun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sylvain Paris](http:\u002F\u002Fpeople.csail.mit.edu\u002Fsparis\u002F)\u003C\u002Fli> \u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli> \u003Cli>[Kavita Bala](https:\u002F\u002Fwww.cs.cornell.edu\u002F~kb\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fluanfujun\u002Fdeep-painterly-harmonization?style=social)](https:\u002F\u002Fgithub.com\u002Fluanfujun\u002Fdeep-painterly-harmonization) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.03189), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1701.08893)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjcjohnson\u002Fneural-style), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftorch\u002Ftorch7), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fszagoruyko\u002Floadcaffe)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgist\u002Feyaler\u002F5303782669fb43510d398bd346c6e3e6\u002Fdeep-painterly-harmonization.ipynb) | 23.09.2024 |\n| CogVideo | Large-scale text-to-video generation model based on diffusion transformer, which can generate 10-second continuous videos aligned with text prompt, with a frame rate of 16 fps and resolution of 768 * 1360 pixels | \u003Cul>\u003Cli>[Zhuoyi Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=tgAt-gEAAAAJ)\u003C\u002Fli> \u003Cli>[Jiayan Teng](https:\u002F\u002Ftengjiayan20.github.io\u002F)\u003C\u002Fli> \u003Cli>[Wendi Zheng](https:\u002F\u002Fgithub.com\u002Fminkowski0125)\u003C\u002Fli> \u003Cli>[Ming Ding](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Va50YzkAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Shiyu Huang](https:\u002F\u002Fgithub.com\u002Fhuangshiyu13)\u003C\u002Fli> \u003Cli>[Jiazheng Xu](https:\u002F\u002Fgithub.com\u002Fxujz18)\u003C\u002Fli> \u003Cli>[Yuanming Yang](https:\u002F\u002Fgithub.com\u002Fyomi1117)\u003C\u002Fli> \u003Cli>[Wenyi Hong](https:\u002F\u002Fgithub.com\u002Fwenyihong)\u003C\u002Fli> \u003Cli>[Xiaohan Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RKyE8o0AAAAJ)\u003C\u002Fli> \u003Cli>[Guanyu Feng](https:\u002F\u002Fgithub.com\u002Fjiguanglizipao)\u003C\u002Fli> \u003Cli>[Da Yin](https:\u002F\u002Fsomefive.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuxuan Zhang](https:\u002F\u002Fgithub.com\u002FzRzRzRzRzRzRzR)\u003C\u002Fli> \u003Cli>[Weihan Wang](https:\u002F\u002Fgithub.com\u002Fmactavish91)\u003C\u002Fli> \u003Cli>[Yean Cheng](https:\u002F\u002Fliquidammonia.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bin Xu](https:\u002F\u002Fkeg.cs.tsinghua.edu.cn\u002Fpersons\u002Fxubin\u002F)\u003C\u002Fli> \u003Cli>[Xiaotao Gu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=YR4Lp0QAAAAJ)\u003C\u002Fli> \u003Cli>[Yuxiao Dong](https:\u002F\u002Fkeg.cs.tsinghua.edu.cn\u002Fyuxiao\u002F)\u003C\u002Fli> \u003Cli>[Jie Tang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=XfFqozoAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTHUDM\u002FCogVideo?style=social)](https:\u002F\u002Fgithub.com\u002FTHUDM\u002FCogVideo) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.06072), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.15868)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fyzy-thu.github.io\u002FCogVideoX-demo\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FdCGfUsagrD)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTHUDM\u002FCogKit), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fa-r-r-o-w\u002Fcogvideox-factory), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsayakpaul\u002Fdiffusers-torchao), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTHUDM\u002FSwissArmyTransformer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fpytorch\u002Fao), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Foptimum-quanto), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Faigc-apps\u002FCogVideoX-Fun), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fthu-ml\u002FRIFLEx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fpinokiofactory\u002Fcogstudio), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkijai\u002FComfyUI-CogVideoXWrapper), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNUS-HPC-AI-Lab\u002FVideoSys), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffeizc\u002FCogvideX-Interpolation), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTheDenk\u002Fcogvideox-controlnet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FVideoVerses\u002FVideoTuna), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxdit-project\u002FxDiT)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FTHUDM\u002FCogVideoX-5B), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FTHUDM\u002FCogVideoX1.5-5B-SAT), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FTHUDM\u002FCogVideoX-5B-Space), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FTHUDM\u002Fcogvlm2-llama3-caption)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1el6uy0\u002Fcogvideox_texttovideo_diffusion_models_with_an\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5UCkMzP2VLE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUD3ZFLj-3uE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FChS4JGijwPk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPPGYNmrVG58), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3IRnXpl9Zmo)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1pCe5s0bC_xuXbBlpvIH1z0kfdTLQPzCS) | 18.09.2024 |\n| audio2photoreal | Framework for generating full-bodied photorealistic avatars that gesture according to the conversational dynamics of a dyadic interaction | \u003Cul>\u003Cli>[Evonne Ng](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~evonne_ng\u002F)\u003C\u002Fli> \u003Cli>[Javier Romero](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Wx62iOsAAAAJ)\u003C\u002Fli> \u003Cli>[Timur Bagautdinov](https:\u002F\u002Fscholar.google.ch\u002Fcitations?user=oLi7xJ0AAAAJ)\u003C\u002Fli> \u003Cli>[Shaojie Bai](https:\u002F\u002Fjerrybai1995.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Trevor Darrell](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~trevor\u002F)\u003C\u002Fli> \u003Cli>[Angjoo Kanazawa](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~kanazawa\u002F)\u003C\u002Fli> \u003Cli>[Alexander Richard](https:\u002F\u002Falexanderrichard.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Faudio2photoreal?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Faudio2photoreal) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.01885)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fca_body)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~evonne_ng\u002Fprojects\u002Faudio2photoreal\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FY0GMaMtUynQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1A6EwKM3PeX7dcKV66zxQWuP-v_dKlX_0) | 13.09.2024 |\n| Fast Segment Anything | CNN Segment Anything Model trained using only 2% of the SA-1B dataset published by SAM authors | \u003Cul>\u003Cli>[Xu Zhao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=F0cYEyAAAAAJ)\u003C\u002Fli> \u003Cli>[Wenchao Ding](https:\u002F\u002Fgithub.com\u002Fberry-ding)\u003C\u002Fli> \u003Cli>[Yongqi An](https:\u002F\u002Fgithub.com\u002Fan-yongqi)\u003C\u002Fli> \u003Cli>[Yinglong Du](https:\u002F\u002Fgithub.com\u002FYinglongDu)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Tao Yu](https:\u002F\u002Fgithub.com\u002Ftianjinren)\u003C\u002Fli> \u003Cli>[Min Li](https:\u002F\u002Fgithub.com\u002Flimin2021)\u003C\u002Fli> \u003Cli>[Ming Tang](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FMing-Tang-2)\u003C\u002Fli> \u003Cli>[Jinqiao Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=7_BkyxEAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCASIA-IVA-Lab\u002FFastSAM?style=social)](https:\u002F\u002Fgithub.com\u002FCASIA-IVA-Lab\u002FFastSAM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.12156), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10003)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FChuRuaNh0\u002FFastSam_Awsome_TensorRT)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@mahimairaja\u002Fso-what-exactly-is-fastsam-the-ultimate-guide-ddae21d3b486)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyHNPyqazYYU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FSslzS0AsiAw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FqvqkjP1wCDE)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1oX14f6IneGGw612WgVlAiy91UHwFAvr9) | 10.09.2024 |\n| Neuralangelo | Framework for high-fidelity 3D surface reconstruction from RGB video captures | \u003Cul>\u003Cli>[Zhaoshuo Li](https:\u002F\u002Fmli0603.github.io\u002F)\u003C\u002Fli> \u003Cli>[Thomas Müller](https:\u002F\u002Ftom94.net\u002F)\u003C\u002Fli> \u003Cli>[Alex Evans](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ToqGImkAAAAJ)\u003C\u002Fli> \u003Cli>[Russell Taylor](https:\u002F\u002Fwww.cs.jhu.edu\u002F~rht\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Mathias Unberath](https:\u002F\u002Fmathiasunberath.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ming-Yu Liu](https:\u002F\u002Fmingyuliu.net\u002F)\u003C\u002Fli> \u003Cli>[Chen-Hsuan Lin](https:\u002F\u002Fchenhsuanlin.bitbucket.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNVlabs\u002Fneuralangelo?style=social)](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fneuralangelo) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.03092)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fblogs.nvidia.com\u002Fblog\u002F2023\u002F06\u002F01\u002Fneuralangelo-ai-research-3d-reconstruction\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmli0603\u002FBlenderNeuralangelo)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fresearch.nvidia.com\u002Flabs\u002Fdir\u002Fneuralangelo\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPQMNCXR-WF8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQpdw3SW54kI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FlC2uPDfaTcE)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1i16s8W_OV0Hd3-PIuo64JKDwwdOesgXQ) | 02.09.2024 |\n| SPIN | Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loop | \u003Cul>\u003Cli>[Nikos Kolotouros](https:\u002F\u002Fwww.nikoskolot.com\u002F)\u003C\u002Fli> \u003Cli>[Georgios Pavlakos](https:\u002F\u002Fgeopavlakos.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli> \u003Cli>[Kostas Daniilidis](https:\u002F\u002Fwww.cis.upenn.edu\u002F~kostas\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_aed0085d183c.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV.2019.00234) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fnkolot\u002FSPIN?style=social)](https:\u002F\u002Fgithub.com\u002Fnkolot\u002FSPIN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.12828)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fchaneyk\u002Fspin)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplify-x), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCMU-Perceptual-Computing-Lab\u002Fopenpose)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.nikoskolot.com\u002Fprojects\u002Fspin\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1uH2JtavOtDrFl6RsipyIncCSr19GWW4x) | 21.08.2024 |\n| YOLOv10 | Aim to further advance the performance-efficiency boundary of YOLOs from both the post-processing and model architecture | \u003Cul>\u003Cli>[Ao Wang](https:\u002F\u002Fgithub.com\u002Fjameslahm)\u003C\u002Fli> \u003Cli>[Hui Chen](https:\u002F\u002Fhuichen24.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kai Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=bZQX708AAAAJ)\u003C\u002Fli> \u003Cli>[Zijia Lin](https:\u002F\u002Fsites.google.com\u002Fsite\u002Flinzijia72)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jungong Han](https:\u002F\u002Fjungonghan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Guiguang Ding](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=B7F3yt4AAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTHU-MIG\u002Fyolov10?style=social)](https:\u002F\u002Fgithub.com\u002FTHU-MIG\u002Fyolov10) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.14458)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Flearnopencv.com\u002Fyolov10\u002F)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fopenbayes.com\u002Fconsole\u002Fpublic\u002Ftutorials\u002Fim29uYrnIoz)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frlggyp\u002FYOLOv10-OpenVINO-CPP-Inference), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FSeeed-Projects\u002Fjetson-examples\u002Fblob\u002Fmain\u002FreComputer\u002Fscripts\u002Fyolov10\u002FREADME.md), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkaylorchen\u002Frk3588-yolo-demo), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenvinotoolkit\u002Fopenvino_notebooks\u002Fblob\u002Flatest\u002Fnotebooks\u002Fyolov10-optimization\u002Fyolov10-optimization.ipynb), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsujanshresstha\u002FYOLOv10_DeepSORT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCVHub520\u002FX-AnyLabeling), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDanielSarmiento04\u002Fyolov10cpp), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flyuwenyu\u002FRT-DETR)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Fjameslahm\u002Fyolov10-665b0d90b0b5bb85129460c2), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fjameslahm\u002FYOLOv10), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fkadirnar\u002FYolov10), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FXenova\u002Fyolov10-web)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@batuhansenerr\u002Fyolov10-custom-object-detection-bd7298ddbfd3), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@sunidhi.ashtekar\u002Fyolov10-revolutionizing-real-time-object-detection-72ef04ad441a)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FGPTFutureScience\u002Fcomments\u002F1d34rj1\u002Fyolov10_the_future_of_realtime_object_detection\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F29tnSxhB3CY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2ZFJbeJXXDM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FwM6nO75keOQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Froboflow-ai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnotebooks\u002Ftrain-yolov10-object-detection-on-custom-dataset.ipynb) | 20.08.2024 |\n| SpecVQGAN | Taming the visually guided sound generation by shrinking a training dataset to a set of representative vectors | \u003Cul>\u003Cli>[Vladimir Iashin](https:\u002F\u002Fiashin.ai\u002F)\u003C\u002Fli> \u003Cli>[Esa Rahtu](https:\u002F\u002Fesa.rahtu.fi\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fv-iashin\u002FSpecVQGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fv-iashin\u002FSpecVQGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F2110.08791), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.09841), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.00937), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.00820), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1712.01393), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1512.08512)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FPeihaoChen\u002Fregnet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftoshas\u002Ftorch-fidelity), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdescriptinc\u002Fmelgan-neurips), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Flyra)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fiashin.ai\u002FSpecVQGAN)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FFoley_(filmmaking)), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FRow-_and_column-major_order), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FKullback%E2%80%93Leibler_divergence)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=Bucb3nAa398)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1pxTIMweAKApJZ3ZFqyBee3HtMqFpnwQ0) | 12.07.2024 |\n| LivePortrait | Video-driven portrait animation framework with a focus on better generalization, controllability, and efficiency for practical usage | \u003Cul>\u003Cli>[Jianzhu Guo](https:\u002F\u002Fguojianzhu.com\u002F)\u003C\u002Fli> \u003Cli>[Dingyun Zhang](https:\u002F\u002Fgithub.com\u002FDingyunZhang)\u003C\u002Fli> \u003Cli>[Xiaoqiang Liu](https:\u002F\u002Fgithub.com\u002FLiu-lxq)\u003C\u002Fli> \u003Cli>[Zhizhou Zhong](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=t88nyvsAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yuan Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=_8k1ubAAAAAJ)\u003C\u002Fli> \u003Cli>[Pengfei Wan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=P6MraaYAAAAJ)\u003C\u002Fli> \u003Cli>[Di Zhang](https:\u002F\u002Fopenreview.net\u002Fprofile?id=~Di_ZHANG3)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FKwaiVGI\u002FLivePortrait?style=social)](https:\u002F\u002Fgithub.com\u002FKwaiVGI\u002FLivePortrait) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.03168)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkijai\u002FComfyUI-LivePortraitKJ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fshadowcz007\u002Fcomfyui-liveportrait), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhanglonghao1992\u002FOne-Shot_Free-View_Neural_Talking_Head_Synthesis), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002FSPADE), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FKwaiVGI\u002FLivePortrait)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fliveportrait.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1dvepjx\u002Fliveportrait_efficient_portrait_animation_with\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuyjSTAOY7yI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8-IcDDmiUMM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FaFcS31OWMjE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbRHf2oQwgG4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FFPtpNrmuwXk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FwG7oPp01COg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FLivePortrait-jupyter\u002Fblob\u002Fmain\u002FLivePortrait_jupyter.ipynb) | 10.07.2024 |\n| Wav2Lip | A Lip Sync Expert Is All You Need for Speech to Lip Generation In the Wild | \u003Cul>\u003Cli>[Prajwal Renukanand](https:\u002F\u002Fgithub.com\u002Fprajwalkr)\u003C\u002Fli> \u003Cli>[Rudrabha Mukhopadhyay](https:\u002F\u002Frudrabha.github.io\u002F)\u003C\u002Fli> \u003Cli>[Vinay Namboodiri](https:\u002F\u002Fvinaypn.github.io\u002F)\u003C\u002Fli> \u003Cli>[C. V. Jawahar](https:\u002F\u002Ffaculty.iiit.ac.in\u002F~jawahar\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2f88fb4d252e.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3394171.3413532) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FRudrabha\u002FWav2Lip?style=social)](https:\u002F\u002Fgithub.com\u002FRudrabha\u002FWav2Lip) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.10010)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vgg\u002Fdata\u002Flip_reading\u002Flrs2.html)\u003C\u002Fli>\u003Cli>[demo](http:\u002F\u002Fbhaasha.iiit.ac.in\u002Flipsync\u002F)\u003C\u002Fli>\u003Cli>[project](http:\u002F\u002Fcvit.iiit.ac.in\u002Fresearch\u002Fprojects\u002Fcvit-projects\u002Fa-lip-sync-expert-is-all-you-need-for-speech-to-lip-generation-in-the-wild\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=0fXaDCZNOJc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Feyaler\u002Favatars4all\u002Fblob\u002Fmaster\u002Fmelaflefon.ipynb) | 27.06.2024 |\n| FELIX | Feature Engineering with LLMs for Interpretability and Explainability, a novel approach harnessing the vast world knowledge embedded in pre-trained Large Language Models to automatically generate a set of features describing the data | \u003Cul>\u003Cli>[Simon Malberg](https:\u002F\u002Fgithub.com\u002Fsimonmalberg)\u003C\u002Fli> \u003Cli>[Edoardo Mosca](https:\u002F\u002Fedoardomosca.github.io\u002F)\u003C\u002Fli> \u003Cli>[Georg Groh](https:\u002F\u002Fsocvm1.cit.tum.de\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_4c82bcaa094d.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-70359-1_14) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsimonmalberg\u002Ffelix?style=social)](https:\u002F\u002Fgithub.com\u002Fsimonmalberg\u002Ffelix)  | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsimonmalberg\u002Ffelix\u002Fblob\u002Fmain\u002FPaper\u002FFELIX.ipynb) | 13.06.2024 |\n| PoolFormer | MetaFormer Is Actually What You Need for Vision | \u003Cul>\u003Cli>[Weihao Yu](https:\u002F\u002Fwhyu.me\u002F)\u003C\u002Fli> \u003Cli>[Mi Luo](https:\u002F\u002Fluomi97.github.io\u002F)\u003C\u002Fli> \u003Cli>[Pan Zhou](https:\u002F\u002Fpanzhous.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chenyang Si](https:\u002F\u002Fgithub.com\u002FChenyangSi)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yichen Zhou](https:\u002F\u002Fdblp.org\u002Fpid\u002F55\u002F10422.html)\u003C\u002Fli> \u003Cli>[Xinchao Wang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fsitexinchaowang\u002F)\u003C\u002Fli> \u003Cli>[Jiashi Feng](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fjshfeng\u002F)\u003C\u002Fli> \u003Cli>[Shuicheng Yan](https:\u002F\u002Fyanshuicheng.ai\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fb8d53f94a8d.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01055) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsail-sg\u002Fpoolformer?style=social)](https:\u002F\u002Fgithub.com\u002Fsail-sg\u002Fpoolformer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.11418)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frwightman\u002Fpytorch-image-models), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffvcore), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fapex)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002Fpoolformer)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsail-sg\u002Fpoolformer\u002Fblob\u002Fmain\u002Fmisc\u002Fpoolformer_demo.ipynb) | 01.06.2024 |\n| StoryDiffusion | Way of self-attention calculation, termed Consistent Self-Attention, that significantly boosts the consistency between the generated images and augments prevalent pretrained diffusion-based text-to-image models in a zero-shot manner | \u003Cul>\u003Cli>[Yupeng Zhou](https:\u002F\u002Fmmcheng.net\u002Fzyp\u002F)\u003C\u002Fli> \u003Cli>[Daquan Zhou](https:\u002F\u002Fgithub.com\u002Fzhoudaquan)\u003C\u002Fli> \u003Cli>[Ming-Ming Cheng](https:\u002F\u002Fmmcheng.net\u002Fcmm\u002F)\u003C\u002Fli> \u003Cli>[Jiashi Feng](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fjshfeng\u002F?pli=1)\u003C\u002Fli> \u003Cli>[Qibin Hou](https:\u002F\u002Fhouqb.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FHVision-NKU\u002FStoryDiffusion?style=social)](https:\u002F\u002Fgithub.com\u002FHVision-NKU\u002FStoryDiffusion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.01434)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGeNyP4VY9rE?si=qW1jcW_GbKutmKQv)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fstorydiffusion.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStoryDiffusion\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjZWRENqCl6I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGeNyP4VY9rE)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FHVision-NKU\u002FStoryDiffusion\u002Fblob\u002Fmain\u002FComic_Generation.ipynb) | 04.05.2024 |\n| FILM | A frame interpolation algorithm that synthesizes multiple intermediate frames from two input images with large in-between motion | \u003Cul>\u003Cli>[Fitsum Reda](https:\u002F\u002Ffitsumreda.github.io\u002F)\u003C\u002Fli> \u003Cli>[Janne Kontkanen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=MnXc4JQAAAAJ)\u003C\u002Fli> \u003Cli>[Eric Tabellion](http:\u002F\u002Fwww.tabellion.org\u002Fet\u002F)\u003C\u002Fli> \u003Cli>[Deqing Sun](https:\u002F\u002Fdeqings.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Caroline Pantofaru](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vKAKE1gAAAAJ)\u003C\u002Fli> \u003Cli>[Brian Curless](https:\u002F\u002Fhomes.cs.washington.edu\u002F~curless\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_d06e71d41283.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20071-7_15) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fframe-interpolation?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fframe-interpolation) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2202.04901)\u003C\u002Fli>\u003Cli>[data](http:\u002F\u002Fdata.csail.mit.edu\u002Ftofu\u002Ftestset\u002Fvimeo_interp_test.zip), [data](https:\u002F\u002Fvision.middlebury.edu\u002Fflow\u002Fdata), [data](https:\u002F\u002Fpeople.cs.umass.edu\u002F~hzjiang\u002Fprojects\u002Fsuperslomo\u002FUCF101_results.zip)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsniklaus\u002Fsoftmax-splatting\u002Fblob\u002Fmaster\u002Fbenchmark.py)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ffilm-net.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Ftutorials\u002Fload_data\u002Ftfrecord), [\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fapi_docs\u002Fpython\u002Ftf\u002Ftrain\u002FExample), [\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fguide\u002Fsaved_model)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FOAD-BieIjH4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1sK0uc-GJxmdnaxHhYqD2afRknakpdTNZ) | 03.05.2024 |\n| VoiceCraft | token infilling neural codec language model, that achieves state-of-the-art performance on both speech editing and zero-shot text-to-speech on audiobooks, internet videos, and podcasts | \u003Cul>\u003Cli>[Puyuan Peng](https:\u002F\u002Fjasonppy.github.io\u002F)\u003C\u002Fli> \u003Cli>[Po-Yao Huang](https:\u002F\u002Fberniebear.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shang-Wen Li](https:\u002F\u002Fswdanielli.github.io\u002F)\u003C\u002Fli> \u003Cli>[Abdelrahman Mohamed](https:\u002F\u002Fwww.cs.toronto.edu\u002F~asamir\u002F)\u003C\u002Fli> \u003Cli>[David Harwath](https:\u002F\u002Fwww.cs.utexas.edu\u002F~harwath\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjasonppy\u002FVoiceCraft?style=social)](https:\u002F\u002Fgithub.com\u002Fjasonppy\u002FVoiceCraft) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.16973)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flifeiteng\u002Fvall-e)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fpyp1\u002FVoiceCraft)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fjasonppy.github.io\u002FVoiceCraft_web\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLocalLLaMA\u002Fcomments\u002F1bmxfk3\u002Fvoicecraft_zeroshot_speech_editing_and\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FeikybOi8iwU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPJ2qSjycLcw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJxRrHpq-hys)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fjasonppy\u002FVoiceCraft\u002Fblob\u002Fmaster\u002Fvoicecraft-gradio-colab.ipynb) | 21.04.2024 |\n| ZeST | Method for zero-shot material transfer to an object in the input image given a material exemplar image | \u003Cul>\u003Cli>[Ta-Ying Cheng](https:\u002F\u002Fttchengab.github.io\u002F)\u003C\u002Fli> \u003Cli>[Prafull Sharma](https:\u002F\u002Fprafullsharma.net\u002F)\u003C\u002Fli> \u003Cli>[Andrew Markham](https:\u002F\u002Fwww.cs.ox.ac.uk\u002Fpeople\u002Fandrew.markham\u002F)\u003C\u002Fli> \u003Cli>[Niki Trigoni](https:\u002F\u002Fwww.cs.ox.ac.uk\u002Fpeople\u002Fniki.trigoni\u002F)\u003C\u002Fli> \u003Cli>[Varun Jampani](https:\u002F\u002Fvarunjampani.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fttchengab\u002Fzest_code?style=social)](https:\u002F\u002Fgithub.com\u002Fttchengab\u002Fzest_code) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.06425)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkealiu\u002FComfyUI-ZeroShot-MTrans)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fh94\u002FIP-Adapter), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fintel-isl\u002FDPT\u002Freleases\u002Fdownload\u002F1_0\u002Fdpt_hybrid-midas-501f0c75.pt)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fxthemadgenius.medium.com\u002Fzest-unlocks-material-magic-in-single-image-transfers-05f7ff7ee483)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fttchengab.github.io\u002Fzest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Flearnmachinelearning\u002Fcomments\u002F1c0wpjd\u002Fzest_zeroshot_material_transfer_from_a_single\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FatG1VvgeG_g)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002Fzest-jupyter\u002Fblob\u002Fmain\u002Fzest_jupyter.ipynb) | 16.04.2024 |\n| InstantMesh | Feed-forward framework for instant 3D mesh generation from a single image, featuring state-of-the-art generation quality and significant training scalability | \u003Cul>\u003Cli>[Jiale Xu](https:\u002F\u002Fgithub.com\u002Fbluestyle97)\u003C\u002Fli> \u003Cli>[Weihao Cheng](https:\u002F\u002Fwww.cheng.website\u002F)\u003C\u002Fli> \u003Cli>[Yiming Gao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=uRCc-McAAAAJ)\u003C\u002Fli> \u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Shenghua Gao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=fe-1v0MAAAAJ)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTencentARC\u002FInstantMesh?style=social)](https:\u002F\u002Fgithub.com\u002FTencentARC\u002FInstantMesh) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.07191)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdanielgatis\u002Frembg), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002F3DTopia\u002FOpenLRM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnv-tlabs\u002FFlexiCubes)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FTencentARC\u002FInstantMesh)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1c5hs3e\u002Finstantmesh_efficient_3d_mesh_generation_from_a\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBvngSJOStvQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FInstantMesh-jupyter\u002Fblob\u002Fmain\u002FInstantMesh_jupyter.ipynb) | 16.04.2024 |\n| Würstchen | Architecture for text-to-image synthesis that combines competitive performance with unprecedented cost-effectiveness for large-scale text-to-image diffusion models | \u003Cul>\u003Cli>[Pablo Pernias](https:\u002F\u002Fgithub.com\u002Fpabloppp)\u003C\u002Fli> \u003Cli>[Dominic Rampas](https:\u002F\u002Fgithub.com\u002Fdome272)\u003C\u002Fli> \u003Cli>[Mats Richter](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=xtlV5SAAAAAJ)\u003C\u002Fli> \u003Cli>[Christopher Pal](https:\u002F\u002Fwww.polymtl.ca\u002Fexpertises\u002Fpal-christopher-j)\u003C\u002Fli> \u003Cli>[Marc Aubreville](https:\u002F\u002Flme.tf.fau.de\u002Fperson\u002Faubreville\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdome272\u002Fwuerstchen?style=social)](https:\u002F\u002Fgithub.com\u002Fdome272\u002Fwuerstchen) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.00637)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fwuerstchen)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F16hsklt\u002Fw%C3%BCrstchen_is_here_a_game_changing_fastest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FogJsCPqgFMk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdome272\u002FWuerstchen\u002Fblob\u002Fmain\u002Fw%C3%BCrstchen-stage-C.ipynb) | 06.04.2024 |\n| BEiT | Self-supervised vision representation model, which stands for Bidirectional Encoder representation from Image Transformers | \u003Cul>\u003Cli>[Hangbo Bao](https:\u002F\u002Faddf400.github.io\u002F)\u003C\u002Fli> \u003Cli>[Li Dong](https:\u002F\u002Fdong.li\u002F)\u003C\u002Fli> \u003Cli>[Songhao Piao](https:\u002F\u002Fhomepage.hit.edu.cn\u002Fpiaosh)\u003C\u002Fli> \u003Cli>[Furu Wei](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Ffuwei\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fc7c1b3c2320.png)](https:\u002F\u002Fdoi.org\u002F10.48550\u002FarXiv.2106.08254) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002Funilm?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Funilm\u002Ftree\u002Fmaster\u002Fbeit) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.10442), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.06366), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.01127)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fapex), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdeit), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fpytorch-image-models), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdinov)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ftransformers\u002Fmaster\u002Fmodel_doc\u002Fbeit.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Freview-beit-bert-pre-training-of-image-transformers-c14a7ef7e295)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fsemantic-segmentation-on-ade20k)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVEEUklIWuNE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FSYEIvkPbyhU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FNielsRogge\u002FTransformers-Tutorials\u002Fblob\u002Fmaster\u002FBEiT\u002FUnderstanding_BeitForMaskedImageModeling.ipynb) | 30.03.2024 |\n| AudioSep | Foundation model for open-domain audio source separation with natural language queries | \u003Cul>\u003Cli>[Xubo Liu](https:\u002F\u002Fliuxubo717.github.io\u002F)\u003C\u002Fli> \u003Cli>[Qiuqiang Kong](https:\u002F\u002Fqiuqiangkong.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yan Zhao](https:\u002F\u002Fcliffzhao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haohe Liu](https:\u002F\u002Fhaoheliu.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yi Yuan](https:\u002F\u002Fwww.surrey.ac.uk\u002Fpeople\u002Fyi-yuan)\u003C\u002Fli> \u003Cli>[Yuzhuo Liu](https:\u002F\u002Fgithub.com\u002Fredrabbit94)\u003C\u002Fli> \u003Cli>[Rui Xia](https:\u002F\u002Fscholar.google.co.uk\u002Fcitations?user=26oErxwAAAAJ)\u003C\u002Fli> \u003Cli>[Yuxuan Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=3RaOfJkAAAAJ)\u003C\u002Fli> \u003Cli>[Mark Plumbley](https:\u002F\u002Fwww.surrey.ac.uk\u002Fpeople\u002Fmark-plumbley)\u003C\u002Fli> \u003Cli>[Wenwu Wang](http:\u002F\u002Fpersonal.ee.surrey.ac.uk\u002FPersonal\u002FW.Wang\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAudio-AGI\u002FAudioSep?style=social)](https:\u002F\u002Fgithub.com\u002FAudio-AGI\u002FAudioSep) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.05037)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Faudio-agi.github.io\u002FSeparate-Anything-You-Describe\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAudio-AGI\u002FAudioSep\u002Fblob\u002Fmain\u002FAudioSep_Colab.ipynb) | 15.03.2024 |\n| AQLM | Extreme Compression of Large Language Models via Additive Quantization | \u003Cul>\u003Cli>[Vage Egiazarian](https:\u002F\u002Fgithub.com\u002FVahe1994)\u003C\u002Fli> \u003Cli>[Andrei Panferov](https:\u002F\u002Fblog.panferov.org\u002F)\u003C\u002Fli> \u003Cli>[Denis Kuznedelev](https:\u002F\u002Fgithub.com\u002FGodofnothing)\u003C\u002Fli> \u003Cli>[Elias Frantar](https:\u002F\u002Fefrantar.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Artem Babenko](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2Kv3JP0AAAAJ)\u003C\u002Fli> \u003Cli>[Dan Alistarh](https:\u002F\u002Fgithub.com\u002Fdalistarh)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FVahe1994\u002FAQLM?style=social)](https:\u002F\u002Fgithub.com\u002FVahe1994\u002FAQLM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.06118)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fdatasets\u002Fmain\u002Fen\u002Fcache#cache-directory), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Ftogethercomputer\u002FRedPajama-Data-1T-Sample), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FVahe1994\u002FAQLM)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLearningMachines\u002Fcomments\u002F1atvrnl\u002F240106118_extreme_compression_of_large_language\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQx8PNk4OkUA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhAHBKAXO-88)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FVahe1994\u002FAQLM\u002Fblob\u002Fmain\u002Fnotebooks\u002Fcolab_example.ipynb) | 08.03.2024 |\n| YOLOv9 | Learning What You Want to Learn Using Programmable Gradient Information | \u003Cul>\u003Cli>[Chien-Yao Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=DkQh4M4AAAAJ)\u003C\u002Fli> \u003Cli>[I-Hau Yeh](https:\u002F\u002Fieeexplore.ieee.org\u002Fauthor\u002F37088448531)\u003C\u002Fli> \u003Cli>[Hong-Yuan Mark Liao](https:\u002F\u002Fhomepage.iis.sinica.edu.tw\u002Fpages\u002Fliao\u002Findex_zh.html)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FWongKinYiu\u002Fyolov9?style=social)](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002Fyolov9) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.13616), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.16921)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Flearnopencv.com\u002Fyolov9-advancing-the-yolo-legacy\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002Fyolor), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FVDIGPKU\u002FDynamicDet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDingXiaoH\u002FRepVGG)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fkadirnar\u002FYolov9), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fmerve\u002Fyolov9)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@Mert.A\u002Fhow-to-use-yolov9-for-object-detection-93598ad88d7d)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FXHT2c8jT3Bc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3iLJ6YWPg28), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fdccf_sJF0Gg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Froboflow-ai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnotebooks\u002Ftrain-yolov9-object-detection-on-custom-dataset.ipynb) | 05.03.2024 |\n| Multi-LoRA Composition | LoRA Switch and LoRA Composite, approaches that aim to surpass traditional techniques in terms of accuracy and image quality, especially in complex compositions | \u003Cul>\u003Cli>[Ming Zhong](https:\u002F\u002Fmaszhongming.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yelong Shen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=S6OFEFEAAAAJ)\u003C\u002Fli> \u003Cli>[Shuohang Wang](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fshuowa\u002F)\u003C\u002Fli> \u003Cli>[Yadong Lu](https:\u002F\u002Fadamlu123.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yizhu Jiao](https:\u002F\u002Fyzjiao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Siru Ouyang](https:\u002F\u002Fozyyshr.github.io\u002F)\u003C\u002Fli> \u003Cli>[Donghan Yu](https:\u002F\u002Fplusross.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiawei Han](https:\u002F\u002Fhanj.cs.illinois.edu\u002F)\u003C\u002Fli> \u003Cli>[Weizhu Chen](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fwzchen\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmaszhongming\u002FMulti-LoRA-Composition?style=social)](https:\u002F\u002Fgithub.com\u002Fmaszhongming\u002FMulti-LoRA-Composition) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.16843)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@letscodeai\u002Fmulti-lora-composition-for-image-generation-f2706528c590)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fninjasaid13\u002Fcomments\u002F1b13q8s\u002Fmultilora_composition_for_image_generation\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FMingZhong_\u002Fstatus\u002F1762347881812443575?s=20)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fmaszhongming.github.io\u002FMulti-LoRA-Composition\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1eSTj6qGOtSY5NaazwwN3meXOzEZxgaZq) | 03.03.2024 |\n| AMARETTO | Multiscale and multimodal inference of regulatory networks to identify cell circuits and their drivers shared and distinct within and across biological systems of human disease | \u003Cul>\u003Cli>[Nathalie Pochet](http:\u002F\u002Fportals.broadinstitute.org\u002Fpochetlab\u002F)\u003C\u002Fli> \u003Cli>[Olivier Gevaert](https:\u002F\u002Fprofiles.stanford.edu\u002Folivier-gevaert)\u003C\u002Fli> \u003Cli>[Mohsen Nabian](https:\u002F\u002Fgithub.com\u002Fmonabiyan)\u003C\u002Fli> \u003Cli>[Jayendra Shinde](https:\u002F\u002Fjayendrashinde91.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Celine Everaert](http:\u002F\u002Fwww.crig.ugent.be\u002Fen\u002Fnode\u002F510)\u003C\u002Fli> \u003Cli>[Thorin Tabor](http:\u002F\u002Fthorin.tabcreations.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgevaertlab\u002FAMARETTO?style=social)](https:\u002F\u002Fgithub.com\u002Fgevaertlab\u002FAMARETTO) \u003Cul>\u003Cli>[bioconductor](https:\u002F\u002Fbioconductor.org\u002Fpackages\u002Frelease\u002Fbioc\u002Fhtml\u002FAMARETTO.html)\u003C\u002Fli>\u003Cli>[project](http:\u002F\u002Fportals.broadinstitute.org\u002Fpochetlab\u002Famaretto.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1JfnRoNgTVX_7VEGAAmjGjwP_yX2tdDxs) | 28.02.2024 |\n| LIDA | Tool for generating grammar-agnostic visualizations and infographics | [Victor Dibia](https:\u002F\u002Fvictordibia.com\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_14e441d2aa9f.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002F2023.acl-demo.11) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002Flida?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Flida) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.02927)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvictordibia\u002Fllmx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flida-project\u002Flida-streamlit)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@c17hawke\u002Flida-automatically-generate-visualization-and-with-llms-the-future-of-data-visualization-6bc556876b46)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmicrosoft.github.io\u002Flida\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FexYi9W-dhME), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FU9K1Cu45nMQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F6xcCwlDx6f8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmicrosoft\u002Flida\u002Fblob\u002Fmain\u002Fnotebooks\u002Ftutorial.ipynb) | 06.02.2024 |\n| ViT | Vision Transformer and MLP-Mixer Architectures | \u003Cul>\u003Cli>[Alexey Dosovitskiy](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=FXNJRDoAAAAJ)\u003C\u002Fli> \u003Cli>[Lucas Beyer](http:\u002F\u002Flucasb.eyer.be)\u003C\u002Fli> \u003Cli>[Alexander Kolesnikov](https:\u002F\u002Fgithub.com\u002Fakolesnikoff)\u003C\u002Fli> \u003Cli>[Dirk Weissenborn](https:\u002F\u002Fgithub.com\u002Fdirkweissenborn)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xiaohua Zhai](https:\u002F\u002Fgithub.com\u002Fxiaohuazhai)\u003C\u002Fli> \u003Cli>[Thomas Unterthiner](https:\u002F\u002Fgithub.com\u002Funtom)\u003C\u002Fli> \u003Cli>[Mostafa Dehghani](https:\u002F\u002Fwww.mostafadehghani.com\u002F)\u003C\u002Fli> \u003Cli>[Matthias Minderer](https:\u002F\u002Fmatthias.minderer.net\u002F)\u003C\u002Fli> \u003Cli>[Georg Heigold](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=WwqlChAAAAAJ)\u003C\u002Fli> \u003Cli>[Sylvain Gelly](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=m7LvuTkAAAAJ)\u003C\u002Fli> \u003Cli>[Jakob Uszkoreit](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=mOG0bwsAAAAJ)\u003C\u002Fli> \u003Cli>[Neil Houlsby](https:\u002F\u002Fneilhoulsby.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fvision_transformer?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fvision_transformer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.11929), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.01601), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.01601), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.10270), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.01548), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.07991), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.08065)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fblog.research.google\u002F2022\u002F04\u002Flocked-image-tuning-adding-language.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fpytorch-image-models), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fflaxformer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fmodels)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@weiwen21\u002Fan-image-is-worth-16x16-words-transformers-for-image-recognition-at-scale-957f88e53726)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTrdevFK_am4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FHZ4j_U3FC94), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F7K4Z8RqjWIk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FoDtcobGQ7xU?si=C2EgZTESzhTXFSq6), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fv6xj_DG-UEo)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Fvision_transformer\u002Fblob\u002Fmain\u002Fvit_jax.ipynb) | 06.02.2024 |\n| Qwen | Comprehensive language model series that encompasses distinct models with varying parameter counts | [qwenlm](https:\u002F\u002Fqwenlm.github.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FQwenLM\u002FQwen?style=social)](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.16609), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.09685), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.14314), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.11088)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FCV4E9rpNSD)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fqwenllm\u002Fqwen)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen2.5), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen-Agent), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FQwenLM\u002Fqwen.cpp), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDao-AILab\u002Fflash-attention), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAutoGPTQ\u002FAutoGPTQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FDeepSpeed)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FQwen)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fpytorch.org\u002Fdocs\u002Fstable\u002Felastic\u002Frun.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fy6Wh4SpRoao), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbJmx_fAOW78), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FALArhCnz8rY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FQwenLM\u002FQwen\u002Fblob\u002Fmaster\u002Frecipes\u002Fquickstart\u002Fqwen.ipynb) | 30.01.2024 |\n| 3D Ken Burns | A reference implementation of 3D Ken Burns Effect from a Single Image using PyTorch - given a single input image, it animates this still image with a virtual camera scan and zoom subject to motion parallax | [Manuel Romero](https:\u002F\u002Fmrm8488.github.io\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ff89b6fc991e.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3355089.3356528) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsniklaus\u002F3d-ken-burns?style=social)](https:\u002F\u002Fgithub.com\u002Fsniklaus\u002F3d-ken-burns) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.05483)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=WrajxHHfRBA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmrm8488\u002Fshared_colab_notebooks\u002Fblob\u002Fmaster\u002F3D_Ken_Burns.ipynb) | 24.01.2024 |\n| VALL-E X | Cross-lingual neural codec language model for cross-lingual speech synthesis | \u003Cul>\u003Cli>[Ziqiang Zhang](https:\u002F\u002Fgithub.com\u002Fonisac-K)\u003C\u002Fli> \u003Cli>[Long Zhou](https:\u002F\u002Flong-zhou.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chengyi Wang](https:\u002F\u002Fcywang97.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sanyuan Chen](https:\u002F\u002Fsanyuan-chen.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yu Wu](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fyuwu1\u002F)\u003C\u002Fli> \u003Cli>[Shujie Liu](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fshujliu\u002F)\u003C\u002Fli> \u003Cli>[Zhuo Chen](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fzhuc\u002F)\u003C\u002Fli> \u003Cli>[Yanqing Liu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=dIJFz4UAAAAJ)\u003C\u002Fli> \u003Cli>[Huaming Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=aJDLg5IAAAAJ)\u003C\u002Fli> \u003Cli>[Jinyu Li](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fjinyli\u002F)\u003C\u002Fli> \u003Cli>[Lei He](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=EKl9yY8AAAAJ)\u003C\u002Fli> \u003Cli>[Sheng Zhao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=689bIIwAAAAJ)\u003C\u002Fli> \u003Cli>[Furu Wei](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Ffuwei\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPlachtaa\u002FVALL-E-X?style=social)](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.03926), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.02111), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.03143)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fplachtaa.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FqCBRmAnTxg)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flifeiteng\u002Fvall-e)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FPlachta\u002FVALL-E-X)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fsyncedreview\u002Fspeak-a-foreign-language-in-your-own-voice-1dafa42f78d9)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fproject\u002Fvall-e-x)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F7qgfoVFQmvk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1yyD_sz531QntLKowMHo-XxorsFBCfKul) | 19.01.2024 |\n| PhotoMaker | Efficient personalized text-to-image generation method, which mainly encodes an arbitrary number of input ID images into a stack ID embedding for preserving ID information | \u003Cul>\u003Cli>[Zhen Li](https:\u002F\u002Fpaper99.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mingdeng Cao](https:\u002F\u002Fgithub.com\u002Fljzycmd)\u003C\u002Fli> \u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhongang Qi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=zJvrrusAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ming-Ming Cheng](https:\u002F\u002Fmmcheng.net\u002Fcmm\u002F)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTencentARC\u002FPhotoMaker?style=social)](https:\u002F\u002Fgithub.com\u002FTencentARC\u002FPhotoMaker) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.04461)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbmaltais\u002FPhotoMaker), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsdbds\u002FPhotoMaker-for-windows), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZHO-ZHO-ZHO\u002FComfyUI-PhotoMaker), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Ffastcomposer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTencentARC\u002FT2I-Adapter), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftencent-ailab\u002FIP-Adapter)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FTencentARC\u002FPhotoMaker)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@christopheverdier\u002Fphotomaker-the-art-of-ai-consistent-characters-generation-cf2cd037bc3e)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fphoto-maker.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F197bfj9\u002Ftencentarc_releases_photomaker\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FNWIdzTEk5O4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FZTck128jfFY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FTencentARC\u002FPhotoMaker\u002Fblob\u002Fmain\u002Fphotomaker_demo.ipynb) | 18.01.2024 |\n| DDColor | End-to-end method with dual decoders for image colorization | \u003Cul>\u003Cli>[Xiaoyang Kang](https:\u002F\u002Fpiddnad.github.io\u002Fxiaoyangkang)\u003C\u002Fli> \u003Cli>[Tao Yang](https:\u002F\u002Fcg.cs.tsinghua.edu.cn\u002Fpeople\u002F~tyang\u002F)\u003C\u002Fli> \u003Cli>[Wenqi Ouyang](https:\u002F\u002Fvicky0522.github.io\u002FWenqi-Ouyang\u002F)\u003C\u002Fli> \u003Cli>[Peiran Ren](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=x5dEuxsAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lingzhi Li](https:\u002F\u002Flingzhili.com\u002F)\u003C\u002Fli> \u003Cli>[Xuansong Xie](https:\u002F\u002Fgithub.com\u002Fxungie)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fpiddnad\u002FDDColor?style=social)](https:\u002F\u002Fgithub.com\u002Fpiddnad\u002FDDColor) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.11613)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjixiaozhong\u002FColorFormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FKIMGEONUNG\u002FBigColor)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FDDColor-colab\u002Fblob\u002Fmain\u002FDDColor_colab.ipynb) | 15.01.2024 |\n| PASD | Pixel-aware stable diffusion network to achieve robust Real-ISR as well as personalized stylization | \u003Cul>\u003Cli>[Tao Yang](https:\u002F\u002Fcg.cs.tsinghua.edu.cn\u002Fpeople\u002F~tyang)\u003C\u002Fli> \u003Cli>[Peiran Ren](http:\u002F\u002Frenpr.org\u002F)\u003C\u002Fli> \u003Cli>[Xuansong Xie](https:\u002F\u002Fgithub.com\u002Fxungie)\u003C\u002Fli> \u003Cli>[Lei Zhang](https:\u002F\u002Fwww4.comp.polyu.edu.hk\u002F~cslzhang)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyangxy\u002FPASD?style=social)](https:\u002F\u002Fgithub.com\u002Fyangxy\u002FPASD) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.14469)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-upscaler-for-automatic1111)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Frunwayml\u002Fstable-diffusion-v1-5), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fnitrosocke\u002Fmo-di-diffusion)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F18qxe5q\u002Fpixelaware_stable_diffusion_for_realistic_image\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1lZ_-rSGcmreLCiRniVT973x6JLjFiC-b) | 12.01.2024 |\n| HandRefiner | Refining Malformed Hands in Generated Images by Diffusion-based Conditional Inpainting | \u003Cul>\u003Cli>[Wenquan Lu](https:\u002F\u002Fgithub.com\u002Fwenquanlu)\u003C\u002Fli> \u003Cli>[Yufei Xu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=hlYWxX8AAAAJ)\u003C\u002Fli> \u003Cli>[Jing Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=9jH5v74AAAAJ)\u003C\u002Fli> \u003Cli>[Chaoyue Wang](https:\u002F\u002Fwang-chaoyue.github.io\u002F)\u003C\u002Fli> \u003Cli>[Dacheng Tao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RwlJNLcAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwenquanlu\u002FHandRefiner?style=social)](https:\u002F\u002Fgithub.com\u002Fwenquanlu\u002FHandRefiner) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.17957)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFannovel16\u002Fcomfyui_controlnet_aux), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMikubill\u002Fsd-webui-controlnet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FMeshGraphormer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1881z4v\u002Fhandrefiner_refining_malformed_hands_in_generated\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTt-Fyn1RA6c)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FHandRefiner-colab\u002Fblob\u002Fmain\u002FHandRefiner_colab.ipynb) | 08.01.2024 |\n| LLaVA | Large Language and Vision Assistant, an end-to-end trained large multimodal model that connects a vision encoder and LLM for general-purpose visual and language understanding | \u003Cul>\u003Cli>[Haotian Liu](https:\u002F\u002Fhliu.cc\u002F)\u003C\u002Fli> \u003Cli>[Chunyuan Li](https:\u002F\u002Fchunyuan.li\u002F)\u003C\u002Fli> \u003Cli>[Qingyang Wu](https:\u002F\u002Fqywu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yong Jae Lee](https:\u002F\u002Fpages.cs.wisc.edu\u002F~yongjaelee\u002F)\u003C\u002Fli> \u003Cli>[Yuheng Li](https:\u002F\u002Fyuheng-li.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhaotian-liu\u002FLLaVA?style=social)](https:\u002F\u002Fgithub.com\u002Fhaotian-liu\u002FLLaVA) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.08485), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.03744), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.00890), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.09958), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.14895)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fllava.hliu.cc\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fggerganov\u002Fllama.cpp\u002Fpull\u002F3436), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FLLaVA-Med), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FUX-Decoder\u002FSegment-Everything-Everywhere-All-At-Once), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FLuodian\u002FOtter), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FInstruction-Tuning-with-GPT-4\u002FGPT-4-LLM)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fliuhaotian\u002FLLaVA-Pretrain), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fliuhaotian\u002FLLaVA-Pretrained-Projectors)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fxthemadgenius.medium.com\u002Fhow-to-use-llava-large-language-and-vision-assistant-732c666b5ed0)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fllava-vl.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FmkI7EPD1vp8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fkx1VpI6JzsY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRxBSmbdJ1I8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FmdYycY4lsuE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Ft7I46dxfmWs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FKRAQkJC-XJU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FLLaVA-colab\u002Fblob\u002Fmain\u002FLLaVA_13b_4bit_vanilla_colab.ipynb) | 22.12.2023 |\n| Background Matting V2 | Real-time, high-resolution background replacement technique which operates at 30fps in 4K resolution, and 60fps for HD on a modern GPU | \u003Cul>\u003Cli>[Shanchuan Lin](https:\u002F\u002Fgithub.com\u002FPeterL1n)\u003C\u002Fli> \u003Cli>[Andrey Ryabtsev](https:\u002F\u002Fgithub.com\u002Fandreyryabtsev)\u003C\u002Fli> \u003Cli>[Soumyadip Sengupta](https:\u002F\u002Fgithub.com\u002Fsenguptaumd)\u003C\u002Fli> \u003Cli>[Brian Curless](https:\u002F\u002Fhomes.cs.washington.edu\u002F~curless\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Steve Seitz](https:\u002F\u002Fwww.smseitz.com\u002F)\u003C\u002Fli> \u003Cli>[Ira Kemelmacher-Shlizerman](https:\u002F\u002Fwww.irakemelmacher.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_27ef0e6ed9d2.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00865) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPeterL1n\u002FBackgroundMattingV2?style=social)](https:\u002F\u002Fgithub.com\u002FPeterL1n\u002FBackgroundMattingV2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.07810)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsenguptaumd\u002FBackground-Matting), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fandreyryabtsev\u002FBGMv2-webcam-plugin-linux)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgrail.cs.washington.edu\u002Fprojects\u002Fbackground-matting-v2\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FoMfPTeYDF9g), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fb7ps21MVyTA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1cTxFq1YuoJ5QPqaTcnskwlHDolnjBkB9) | 22.12.2023 |\n| FreeInit | Concise yet effective method to improve temporal consistency of videos generated by diffusion modelsconcise yet effective method to improve temporal consistency of videos generated by diffusion models | \u003Cul>\u003Cli>[Tianxing Wu](https:\u002F\u002Ftianxingwu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chenyang Si](http:\u002F\u002Fchenyangsi.top\u002F)\u003C\u002Fli> \u003Cli>[Yuming Jiang](https:\u002F\u002Fyumingj.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ziqi Huang](https:\u002F\u002Fziqihuangg.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cba6c3f73678.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-72646-0_22) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTianxingWu\u002FFreeInit?style=social)](https:\u002F\u002Fgithub.com\u002FTianxingWu\u002FFreeInit) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.07537)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FKosinkadink\u002FComfyUI-AnimateDiff-Evolved)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FTianxingWu\u002FFreeInit)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftianxingwu.github.io\u002Fpages\u002FFreeInit\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FlS5IYbAqriI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FFreeInit-colab\u002Fblob\u002Fmain\u002FFreeInit_colab.ipynb) | 21.12.2023 |\n| Gaussian Splatting | State-of-the-art visual quality while maintaining competitive training times and importantly allow high-quality real-time (≥ 100 fps) novel-view synthesis at 1080p resolution | \u003Cul>\u003Cli>[Bernhard Kerbl](https:\u002F\u002Fwww.cg.tuwien.ac.at\u002Fstaff\u002FBernhardKerbl)\u003C\u002Fli> \u003Cli>[Georgios Kopanas](https:\u002F\u002Fgrgkopanas.github.io\u002F)\u003C\u002Fli> \u003Cli>[Thomas Leimkühler](https:\u002F\u002Fpeople.mpi-inf.mpg.de\u002F~tleimkue\u002F)\u003C\u002Fli> \u003Cli>[George Drettakis](http:\u002F\u002Fwww-sop.inria.fr\u002Fmembers\u002FGeorge.Drettakis\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0c34c9f6bf85.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3592433) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgraphdeco-inria\u002Fgaussian-splatting?style=social)](https:\u002F\u002Fgithub.com\u002Fgraphdeco-inria\u002Fgaussian-splatting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.04079)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcamenduru\u002Fgaussian-splatting)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Faxinc-ai\u002F3d-gaussian-splatting-real-time-rendering-of-photorealistic-scenes-f7f1a47f060)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Frepo-sam.inria.fr\u002Ffungraph\u002F3d-gaussian-splatting\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fsingularity\u002Fcomments\u002F163jeqa\u002F3d_gaussian_splatting_for_realtime_radiance_field\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FT_kXY43VZnk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUXtuigy_wYc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FHVv_IQKlafQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fw43KV79LsFw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTLK3TDDcJFU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FkShNYOuDnlI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjuRMRej2d5c)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002Fgaussian-splatting-colab\u002Fblob\u002Fmain\u002Fgaussian_splatting_colab.ipynb) | 19.12.2023 |\n| SMPLer-X | Scaling up EHPS towards the first generalist foundation model, with up to ViT-Huge as the backbone and training with up to 4.5M instances from diverse data sources | \u003Cul>\u003Cli>[Zhongang Cai](https:\u002F\u002Fcaizhongang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Wanqi Yin](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=zlIJwBEAAAAJ)\u003C\u002Fli> \u003Cli>[Ailing Zeng](https:\u002F\u002Failingzeng.site\u002F)\u003C\u002Fli> \u003Cli>[Chen Wei](https:\u002F\u002Fgithub.com\u002FWei-Chen-hub)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Qingping Sun](https:\u002F\u002Fgithub.com\u002Fttxskk)\u003C\u002Fli> \u003Cli>[Yanjun Wang](https:\u002F\u002Fgithub.com\u002FWYJSJTU)\u003C\u002Fli> \u003Cli>[Hui En Pang](https:\u002F\u002Fpangyyyyy.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haiyi Mei](https:\u002F\u002Fhaiyi-mei.com\u002F)\u003C\u002Fli> \u003Cli>[Mingyuan Zhang](https:\u002F\u002Fmingyuan-zhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Lei Zhang](https:\u002F\u002Fwww.leizhang.org\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli> \u003Cli>[Lei Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jZH2IPYAAAAJ)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcaizhongang\u002FSMPLer-X?style=social)](https:\u002F\u002Fgithub.com\u002Fcaizhongang\u002FSMPLer-X) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.17448)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopen-mmlab\u002Fmmhuman3d\u002Fblob\u002Fmain\u002Fdocs\u002Fhuman_data.md), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmks0601\u002FHand4Whole_RELEASE), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002FOSX)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fneurips.cc\u002Fvirtual\u002F2023\u002Fposter\u002F73473)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fcaizhongang.com\u002Fprojects\u002FSMPLer-X\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fmachinelearningnews\u002Fcomments\u002F176c5z7\u002Fthis_ai_research_proposes_smplerx_a_generalist\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDepTqbPpVzY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FaFTGFInUnM4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FSMPLer-X-colab\u002Fblob\u002Fmain\u002FSMPLer_X_colab.ipynb) | 18.12.2023 |\n| DeepCache | Training-free paradigm that accelerates diffusion models from the perspective of model architecture | \u003Cul>\u003Cli>[Xinyin Ma](https:\u002F\u002Fhorseee.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gongfan Fang](https:\u002F\u002Ffangggf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xinchao Wang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fsitexinchaowang\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhorseee\u002FDeepCache?style=social)](https:\u002F\u002Fgithub.com\u002Fhorseee\u002FDeepCache) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.00858)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fdiffusers\u002Fv0.24.0\u002Fen\u002Fapi\u002Fpipelines\u002Fstable_diffusion\u002Ftext2img#diffusers.StableDiffusionPipeline)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fhorseee.github.io\u002FDiffusion_DeepCache\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F18b40hh\u002Fdeepcache_accelerating_diffusion_models_for_free\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FDeepCache-colab\u002Fblob\u002Fmain\u002FDeepCache_colab.ipynb) | 18.12.2023 |\n| MagicAnimate | Diffusion-based framework that aims at enhancing temporal consistency, preserving reference image faithfully, and improving animation fidelity | \u003Cul>\u003Cli>[Zhongcong Xu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=-4iADzMAAAAJ)\u003C\u002Fli> \u003Cli>[Jianfeng Zhang](http:\u002F\u002Fjeff95.me\u002F)\u003C\u002Fli> \u003Cli>[Jun Hao Liew](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=8gm-CYYAAAAJ)\u003C\u002Fli> \u003Cli>[Hanshu Yan](https:\u002F\u002Fhanshuyan.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jiawei Liu](https:\u002F\u002Fjia-wei-liu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chenxu Zhang](https:\u002F\u002Fzhangchenxu528.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiashi Feng](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fjshfeng\u002Fhome)\u003C\u002Fli> \u003Cli>[Mike Shou](https:\u002F\u002Fsites.google.com\u002Fview\u002Fshowlab)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagic-research\u002Fmagic-animate?style=social)](https:\u002F\u002Fgithub.com\u002Fmagic-research\u002Fmagic-animate) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.16498)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fzcxu-eric\u002FMagicAnimate), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Frunwayml\u002Fstable-diffusion-v1-5), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fstabilityai\u002Fsd-vae-ft-mse)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@AIWorldBlog\u002Frevolutionizing-image-animation-with-magicanimate-technology-78cc94151915)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fshowlab.github.io\u002Fmagicanimate\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fwww.magicanimate.org\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Ftd27SyA9M80), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F1pATjLFvNtY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FHeXknItbMM8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FMagicAnimate-colab\u002Fblob\u002Fmain\u002FMagicAnimate_colab.ipynb) | 18.12.2023 |\n| DiffBIR | Towards Blind Image Restoration with Generative Diffusion Prior | \u003Cul>\u003Cli>[Xinqi Lin](https:\u002F\u002Fgithub.com\u002F0x3f3f3f3fun)\u003C\u002Fli> \u003Cli>[Jingwen He](https:\u002F\u002Fgithub.com\u002Fhejingwenhejingwen)\u003C\u002Fli> \u003Cli>[Ziyan Chen](https:\u002F\u002Fgithub.com\u002Fziyannchen)\u003C\u002Fli> \u003Cli>[Zhaoyang Lyu](https:\u002F\u002Fzhaoyanglyu.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ben Fei](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=skQROj8AAAAJ)\u003C\u002Fli> \u003Cli>[Bo Dai](http:\u002F\u002Fdaibo.info\u002F)\u003C\u002Fli> \u003Cli>[Wanli Ouyang](https:\u002F\u002Fwlouyang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yu Qiao](https:\u002F\u002Fmmlab.siat.ac.cn\u002Fyuqiao)\u003C\u002Fli> \u003Cli>[Chao Dong](http:\u002F\u002Fxpixel.group\u002F2010\u002F01\u002F20\u002Fchaodong.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FXPixelGroup\u002FDiffBIR?style=social)](https:\u002F\u002Fgithub.com\u002FXPixelGroup\u002FDiffBIR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.15070)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Falbarji\u002Fmixture-of-diffusers)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fstabilityai\u002Fstable-diffusion-2-1-base)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002F0x3f3f3f3fun.github.io\u002Fprojects\u002Fdiffbir\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FrGnrpxWjBOg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FMIRiJGuGqsg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FDiffBIR-colab\u002Fblob\u002Fmain\u002FDiffBIR_colab.ipynb) | 18.12.2023 |\n| Segment and Track Anything | Framewoork that allows users to precisely and effectively segment and track any object in a video | \u003Cul>\u003Cli>[Yangming Cheng](https:\u002F\u002Fgithub.com\u002Fyamy-cheng)\u003C\u002Fli> \u003Cli>[Liulei Li](https:\u002F\u002Fgithub.com\u002FlingorX)\u003C\u002Fli> \u003Cli>[Yuanyou Xu](https:\u002F\u002Fgithub.com\u002Fyoxu515)\u003C\u002Fli> \u003Cli>[Xiaodi Li](https:\u002F\u002Fgithub.com\u002FLiNO3Dy)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Zongxin Yang](https:\u002F\u002Fz-x-yang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Wenguan Wang](https:\u002F\u002Fsites.google.com\u002Fview\u002Fwenguanwang)\u003C\u002Fli> \u003Cli>[Yi Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RMSuNFwAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fz-x-yang\u002FSegment-and-Track-Anything?style=social)](https:\u002F\u002Fgithub.com\u002Fz-x-yang\u002FSegment-and-Track-Anything) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.06558), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.02643), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002Feb890c36af87e4ca82e8ef7bcba6a284-Abstract-Conference.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyoxu515\u002Faot-benchmark)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FShilongLiu\u002FGroundingDINO\u002Fresolve\u002Fmain\u002Fgroundingdino_swint_ogc.pth)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002Feb890c36af87e4ca82e8ef7bcba6a284-Abstract-Conference.html), [\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2021\u002Fhash\u002F147702db07145348245dc5a2f2fe5683-Abstract.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDF0iFSsX8KY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUJvKPng9_DA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fm1oFavjIaCM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUPhtpf1k6HA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FXyd54AngvV8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FeZrdna8JkoQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhPjw28Ul4cw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fl7hXM1a3nEA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FnXfq17X6ohk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5oieHqFIJPc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FcK5MPFdJdSY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUFtwFaOfx2I)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1R10N70AJaslzADFqb-a5OihYkllWEVxB) | 08.12.2023 |\n| AudioLDM | Text-to-audio system that is built on a latent space to learn the continuous audio representations from contrastive language-audio pretraining latents | \u003Cul>\u003Cli>[Haohe Liu](https:\u002F\u002Fhaoheliu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zehua Chen](https:\u002F\u002Fgithub.com\u002FzehuachenImperial)\u003C\u002Fli> \u003Cli>[Yi Yuan](https:\u002F\u002Fwww.surrey.ac.uk\u002Fpeople\u002Fyi-yuan)\u003C\u002Fli> \u003Cli>[Xinhao Mei](https:\u002F\u002Fxinhaomei.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xubo Liu](https:\u002F\u002Fliuxubo717.github.io\u002F)\u003C\u002Fli> \u003Cli>[Danilo Mandic](https:\u002F\u002Fwww.imperial.ac.uk\u002Fpeople\u002Fd.mandic)\u003C\u002Fli> \u003Cli>[Wenwu Wang](http:\u002F\u002Fpersonal.ee.surrey.ac.uk\u002FPersonal\u002FW.Wang\u002F)\u003C\u002Fli> \u003Cli>[Mark Plumbley](https:\u002F\u002Fwww.surrey.ac.uk\u002Fpeople\u002Fmark-plumbley)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhaoheliu\u002FAudioLDM?style=social)](https:\u002F\u002Fgithub.com\u002Fhaoheliu\u002FAudioLDM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.12503)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FLAION-AI\u002FCLAP), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCompVis\u002Fstable-diffusion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftoshas\u002Ftorch-fidelity)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Faudioldm.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_0VTltNYhao)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Folaviinha\u002FNeuralTextToAudio\u002Fblob\u002Fmain\u002FAudioLDM_pub.ipynb) | 02.12.2023 |\n| TabPFN | Neural network that learned to do tabular data prediction | \u003Cul>\u003Cli>[Noah Hollmann](https:\u002F\u002Fgithub.com\u002Fnoahho)\u003C\u002Fli> \u003Cli>[Samuel Müller](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=pevYEjAAAAAJ)\u003C\u002Fli> \u003Cli>[Katharina Eggensperger](https:\u002F\u002Fgithub.com\u002FKEggensperger)\u003C\u002Fli> \u003Cli>[Frank Hutter](https:\u002F\u002Fml.informatik.uni-freiburg.de\u002Fprofile\u002Fhutter\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fautoml\u002FTabPFN?style=social)](https:\u002F\u002Fgithub.com\u002Fautoml\u002FTabPFN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.01848), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.11189), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.01342), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.03253), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.11189), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10510)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fwww.automl.org\u002Ftabpfn-a-transformer-that-solves-small-tabular-classification-problems-in-a-second\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002Ftunguz\u002Fstatus\u002F1578730907711655937)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBGTO5N5-ack)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F194mCs6SEPEW6C0rcP7xWzcEtt1RBc8jJ) | 29.11.2023 |\n| Concept Sliders | Plug-and-play low rank adaptors applied on top of pretrained models | \u003Cul>\u003Cli>[Rohit Gandikota](https:\u002F\u002Frohitgandikota.github.io\u002F)\u003C\u002Fli> \u003Cli>[Joanna Materzyńska](https:\u002F\u002Fjoaanna.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tingrui Zhou](https:\u002F\u002Fwww.p1at.dev\u002F)\u003C\u002Fli> \u003Cli>[Antonio Torralba](https:\u002F\u002Fgroups.csail.mit.edu\u002Fvision\u002Ftorralbalab\u002F)\u003C\u002Fli> \u003Cli>[David Bau](https:\u002F\u002Fbaulab.info\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Frohitgandikota\u002Fsliders?style=social)](https:\u002F\u002Fgithub.com\u002Frohitgandikota\u002Fsliders) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.12092), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.12598)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@furkangozukara\u002Fconcept-sliders-lora-adaptors-for-precise-control-in-diffusion-models-b7f6b36fabee)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2020\u002Fhash\u002F49856ed476ad01fcff881d57e161d73f-Abstract.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsliders.baulab.info\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F180zon7\u002Fconcept_sliders_lora_adaptors_for_precise_control\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Frohitgandikota\u002Fsliders\u002Fblob\u002Fmain\u002Fdemo_concept_sliders.ipynb) | 26.11.2023 |\n| Qwen-VL | Set of large-scale vision-language models designed to perceive and understand both text and images | \u003Cul>\u003Cli>[Jinze Bai](https:\u002F\u002Fgithub.com\u002Fjinze1994)\u003C\u002Fli> \u003Cli>[Shuai Bai](https:\u002F\u002Fgithub.com\u002FShuaiBai623)\u003C\u002Fli> \u003Cli>[Shusheng Yang](https:\u002F\u002Fshushengyang.com\u002F)\u003C\u002Fli> \u003Cli>[Shijie Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=DuAqyTwAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Sinan Tan](https:\u002F\u002Fwww.tinytangent.com\u002F)\u003C\u002Fli> \u003Cli>[Peng Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=7fjqA0YAAAAJ)\u003C\u002Fli> \u003Cli>[Junyang Lin](https:\u002F\u002Fjustinlin610.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chang Zhou](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QeSoG3sAAAAJ)\u003C\u002Fli> \u003Cli>[Jingren Zhou](http:\u002F\u002Fwww.cs.columbia.edu\u002F~jrzhou\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FQwenLM\u002FQwen-VL?style=social)](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen-VL) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.12966), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.09685), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.14314)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fmodelscope.cn\u002Fstudios\u002Fqwen\u002FQwen-VL-Chat-Demo\u002Fsummary)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002Fz3GAxXZ9Ce)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FBradyFU\u002FAwesome-Multimodal-Large-Language-Models\u002Ftree\u002FEvaluation), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOFA-Sys\u002FTouchStone), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FPanQiWei\u002FAutoGPTQ)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FAILab-CVC\u002FSEED-Bench_Leaderboard), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FQwen\u002FQwen-VL)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FElrSJDg23Po), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FE3MS8GfGWj4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fju09YaO7BGA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FQwen-VL-Chat-colab\u002Fblob\u002Fmain\u002FQwen_VL_Chat_colab.ipynb) | 24.11.2023 |\n| AnimeGANv3 | Double-tail generative adversarial network for fast photo animation | \u003Cul>\u003Cli>[Gang Liu](https:\u002F\u002Fgithub.com\u002Flg0061408)\u003C\u002Fli> \u003Cli>[Xin Chen](https:\u002F\u002Fgithub.com\u002FTachibanaYoshino)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c2bb7f34e2f5.png)](http:\u002F\u002Fdoi.org\u002F10.1587\u002Ftransinf.2023EDP7061) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTachibanaYoshino\u002FAnimeGANv3?style=social)](https:\u002F\u002Fgithub.com\u002FTachibanaYoshino\u002FAnimeGANv3) \u003Cul>\u003Cli>[project](https:\u002F\u002Ftachibanayoshino.github.io\u002FAnimeGANv3\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FEosubeJmAnE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5qLUflWb45E), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FiFjiaPlhVm4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FvJqQQMRYKh0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0KaScDxgyBw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F6WXhjXb5a-o)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1XYNWwM8Xq-U7KaTOqNap6A-Yq1f-V-FB) | 23.11.2023 |\n| Ithaca | First Deep Neural Network for the textual restoration, geographical and chronological attribution of ancient Greek inscriptions | \u003Cul>\u003Cli>[Yannis Assael](https:\u002F\u002Fwww.assael.gr\u002F)\u003C\u002Fli> \u003Cli>[Thea Sommerschield](https:\u002F\u002Ftheasommerschield.it\u002F)\u003C\u002Fli> \u003Cli>[Brendan Shillingford](https:\u002F\u002Fgithub.com\u002Fbshillingford)\u003C\u002Fli> \u003Cli>[Mahyar Bordbar](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=KB3ldSQAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[John Pavlopoulos](https:\u002F\u002Fipavlopoulos.github.io\u002F)\u003C\u002Fli> \u003Cli>[Marita Chatzipanagiotou](https:\u002F\u002Fgr.linkedin.com\u002Fin\u002Fmarita-chatzipanagiotou-b0611a1a2)\u003C\u002Fli> \u003Cli>[Ion Androutsopoulos](https:\u002F\u002Fpages.aueb.gr\u002Fusers\u002Fion\u002F)\u003C\u002Fli> \u003Cli>[Jonathan Prag](https:\u002F\u002Fwww.classics.ox.ac.uk\u002Fpeople\u002Fdr-jonathan-prag)\u003C\u002Fli> \u003Cli>[Nando de Freitas](https:\u002F\u002Fwww.cs.ox.ac.uk\u002Fpeople\u002Fnando.defreitas\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cac39cffe76a.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41586-022-04448-z) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Fithaca?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fithaca) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.06262)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsommerschield\u002Fiphi)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fodsc.medium.com\u002Fdeep-neural-networks-could-be-key-to-ancient-text-restoration-and-attribution-research-shows-81a2d89d9413), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fsyncedreview\u002Fithaca-paper-published-in-nature-the-first-dnn-designed-for-textual-restoration-and-geographical-ef395d56697e)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fithaca.deepmind.com\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Ftgeo0q\u002Fr_restoring_and_attributing_ancient_texts_using\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Fithaca\u002Fblob\u002Fmaster\u002Fcolabs\u002Fithaca_inference.ipynb) | 21.11.2023 |\n| PixArt-Σ | Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation | \u003Cul>\u003Cli>[Junsong Chen](https:\u002F\u002Flawrence-cj.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chongjian Ge](https:\u002F\u002Fchongjiange.github.io\u002F)\u003C\u002Fli> \u003Cli>[Enze Xie](https:\u002F\u002Fxieenze.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yue Wu](https:\u002F\u002Fyuewuhkust.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lewei Yao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=hqDyTg8AAAAJ)\u003C\u002Fli> \u003Cli>[Xiaozhe Ren](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=3t2j87YAAAAJ)\u003C\u002Fli> \u003Cli>[Zhongdao Wang](https:\u002F\u002Fzhongdao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ping Luo](http:\u002F\u002Fluoping.me\u002F)\u003C\u002Fli> \u003Cli>[Huchuan Lu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=D3nE0agAAAAJ)\u003C\u002Fli> \u003Cli>[Zhenguo Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=XboZC1AAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPixArt-alpha\u002FPixArt-sigma?style=social)](https:\u002F\u002Fgithub.com\u002FPixArt-alpha\u002FPixArt-sigma) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.04692), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.00426), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.05252)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002Frde6eaE5Ta)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FPixArt-alpha\u002FPixArt-alpha), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FPixArt-alpha\u002FPixArt-LCM)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpixart-alpha.github.io\u002FPixArt-sigma-project\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FPixArtSigma\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1jZ5UZXk7tcpTfVwnX33dDuefNMcnW9ME) | 07.11.2023 |\n| Zero123++ | Image-conditioned diffusion model for generating 3D-consistent multi-view images from a single input view | \u003Cul>\u003Cli>[Ruoxi Shi](https:\u002F\u002Frshi.top\u002F)\u003C\u002Fli> \u003Cli>[Hansheng Chen](https:\u002F\u002Flakonik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhuoyang Zhang](https:\u002F\u002Fgithub.com\u002Fzhuoyang20)\u003C\u002Fli> \u003Cli>[Minghua Liu](https:\u002F\u002Fcseweb.ucsd.edu\u002F~mil070\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chao Xu](https:\u002F\u002Fchaoxu.xyz\u002F)\u003C\u002Fli> \u003Cli>[Xinyue Wei](https:\u002F\u002Fsarahweiii.github.io\u002F)\u003C\u002Fli> \u003Cli>[Linghao Chen](https:\u002F\u002Footts.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chong Zeng](https:\u002F\u002Fwww.chong-zeng.com\u002F)\u003C\u002Fli> \u003Cli>[Hao Su](https:\u002F\u002Fcseweb.ucsd.edu\u002F~haosu\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FSUDO-AI-3D\u002Fzero123plus?style=social)](https:\u002F\u002Fgithub.com\u002FSUDO-AI-3D\u002Fzero123plus) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.15110)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOne-2-3-45\u002FOne-2-3-45), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcvlab-columbia\u002Fzero123)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fsudo-ai\u002Fzero123plus-demo-space), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fysharma\u002FZero123PlusDemo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fxthemadgenius.medium.com\u002Fzero123-your-guide-to-single-view-to-multi-view-3d-image-transformation-b4346b0e6615)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F17f4c6p\u002Fzero123_a_single_image_to_consistent_multiview\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FV9AR-81pAgk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1_5ECnTOosRuAsm2tUp0zvBG0DppL-F3V) | 26.10.2023 |\n| UniFormerV2 | Unified Transformer for Efficient Spatiotemporal Representation Learning | \u003Cul>\u003Cli>[Kunchang Li](https:\u002F\u002Fgithub.com\u002FAndy1621)\u003C\u002Fli> \u003Cli>[Yali Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=hD948dkAAAAJ)\u003C\u002Fli> \u003Cli>[Yinan He](https:\u002F\u002Fgithub.com\u002Fyinanhe)\u003C\u002Fli> \u003Cli>[Yizhuo Li](http:\u002F\u002Fliyizhuo.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yi Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Xm2M8UwAAAAJ)\u003C\u002Fli> \u003Cli>[Limin Wang](http:\u002F\u002Fwanglimin.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yu Qiao](http:\u002F\u002Fmmlab.siat.ac.cn\u002Fyuqiao\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cba6c3f73678.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.00157) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FOpenGVLab\u002FUniFormerV2?style=social)](https:\u002F\u002Fgithub.com\u002FOpenGVLab\u002FUniFormerV2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.09552)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Finnat\u002FUniFormerV2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FAndy1621\u002Funiformerv2_demo), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fpytorch-image-models\u002Fblob\u002Fmain\u002Ftimm\u002Fmodels\u002Fvision_transformer.py), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FSlowFast)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FAndy1621\u002Funiformerv2_demo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Faction-classification-on-activitynet?p=uniformerv2-spatiotemporal-learning-by-arming), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Faction-recognition-on-hacs?p=uniformerv2-spatiotemporal-learning-by-arming), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Faction-classification-on-moments-in-time?p=uniformerv2-spatiotemporal-learning-by-arming), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Faction-recognition-in-videos-on-something-1?p=uniformerv2-spatiotemporal-learning-by-arming), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Faction-classification-on-kinetics-700?p=uniformerv2-spatiotemporal-learning-by-arming)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Z6RzLcno_eLGD_E96mlWoyGieGbKxPQr) | 20.10.2023 |\n| Show-1 | Hybrid model, dubbed as Show-1, which marries pixel-based and latent-based VDMs for text-to-video generation | \u003Cul>\u003Cli>[David Junhao Zhang](https:\u002F\u002Fjunhaozhang98.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jay Zhangjie Wu](https:\u002F\u002Fzhangjiewu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiawei Liu](https:\u002F\u002Fjia-wei-liu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Rui Zhao](https:\u002F\u002Fruizhaocv.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lingmin Ran](https:\u002F\u002Fsiacorplab.nus.edu.sg\u002Fpeople\u002Fran-lingmin\u002F)\u003C\u002Fli> \u003Cli>[Yuchao Gu](https:\u002F\u002Fycgu.site\u002F)\u003C\u002Fli> \u003Cli>[Difei Gao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=No9OsocAAAAJ)\u003C\u002Fli> \u003Cli>[Mike Zheng Shou](https:\u002F\u002Fsites.google.com\u002Fview\u002Fshowlab\u002Fhome)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fshowlab\u002FShow-1?style=social)](https:\u002F\u002Fgithub.com\u002Fshowlab\u002FShow-1) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.15818)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fshowlab\u002Fshow-1-base), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fshowlab\u002Fshow-1-interpolation), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fshowlab\u002Fshow-1-sr1), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fshowlab\u002Fshow-1-sr2), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdamo-vilab\u002Fmodelscope-damo-text-to-video-synthesis), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcerspense\u002Fzeroscope_v2_576w)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fshowlab.github.io\u002FShow-1\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FShow-1-colab\u002Fblob\u002Fmain\u002FShow_1_steps_colab.ipynb) | 15.10.2023 |\n| DA-CLIP | Degradation-aware vision-language model to better transfer pretrained vision-language models to low-level vision tasks as a universal framework for image restoration | \u003Cul>\u003Cli>[Ziwei Luo](https:\u002F\u002Falgolzw.github.io\u002F)\u003C\u002Fli> \u003Cli>[Fredrik Gustafsson](http:\u002F\u002Fwww.fregu856.com\u002F)\u003C\u002Fli> \u003Cli>[Zheng Zhao](https:\u002F\u002Fzz.zabemon.com\u002F)\u003C\u002Fli> \u003Cli>[Jens Sjölund](https:\u002F\u002Fgithub.com\u002Fjsjol)\u003C\u002Fli> \u003Cli>[Thomas Schön](https:\u002F\u002Fuser.it.uu.se\u002F~thosc112\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAlgolzw\u002Fdaclip-uir?style=social)](https:\u002F\u002Fgithub.com\u002FAlgolzw\u002Fdaclip-uir) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.01018)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAlgolzw\u002Fimage-restoration-sde)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fweblzw\u002Fdaclip-uir-ViT-B-32-irsde)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Falgolzw.github.io\u002Fdaclip-uir\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002Fdaclip-uir-colab\u002Fblob\u002Fmain\u002Fdaclip_uir_gradio_colab.ipynb) | 11.10.2023 |\n| SadTalker | Generates 3D motion coefficients of the 3DMM from audio and implicitly modulates a novel 3D-aware face render for talking head generation | \u003Cul>\u003Cli>[Wenxuan Zhang](https:\u002F\u002Fgithub.com\u002FWinfredy)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002Facademic\u002F)\u003C\u002Fli> \u003Cli>[Xuan Wang](https:\u002F\u002Fxuanwangvc.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yong Zhang](https:\u002F\u002Fyzhang2016.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xi Shen](https:\u002F\u002Fxishen0220.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yu Guo](https:\u002F\u002Fyuguo-xjtu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli> \u003Cli>[Fei Wang](http:\u002F\u002Fgr.xjtu.edu.cn\u002Fzh\u002Fweb\u002Ffeynmanw)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_125d6c68e8af.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52729.2023.00836) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FOpenTalker\u002FSadTalker?style=social)](https:\u002F\u002Fgithub.com\u002FOpenTalker\u002FSadTalker) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.12194)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FrrayYqZ4tf)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOpenTalker\u002FDPE), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhanglonghao1992\u002FOne-Shot_Free-View_Neural_Talking_Head_Synthesis), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FRenYurui\u002FPIRender), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FDeep3DFaceReconstruction), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002Ffacexlib), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZz-ww\u002FSadTalker-Video-Lip-Sync), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFeiiYin\u002FSPI)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsadtalker.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAoIzJWnQw1M), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfDgQcDL-qOc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBkSnM9cxkcM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F7u0FYVPQ5rc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FOpenTalker\u002FSadTalker\u002Fblob\u002Fmain\u002Fquick_demo.ipynb) | 10.10.2023 |\n| Musika | Music generation system that can be trained on hundreds of hours of music using a single consumer GPU, and that allows for much faster than real-time generation of music of arbitrary length on a consumer CPU | \u003Cul>\u003Cli>[Marco Pasini](https:\u002F\u002Fgithub.com\u002Fmarcoppasini)\u003C\u002Fli> \u003Cli>[Jan Schlüter](https:\u002F\u002Fwww.ofai.at\u002F~jan.schlueter\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmarcoppasini\u002Fmusika?style=social)](https:\u002F\u002Fgithub.com\u002Fmarcoppasini\u002Fmusika) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.08706), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.08526)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fmagenta.tensorflow.org\u002Fdatasets\u002Fmaestro)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhendriks73\u002Ftempo-cnn), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCPJKU\u002Fmadmom)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fmarcop\u002Fmusika)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmarcoppasini.github.io\u002Fmusika)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQBl8y2Z_i7Y), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0l7OSM-bFvc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1PowSw3doBURwLE-OTCiWkO8HVbS5paRb) | 09.10.2023 |\n| YOLOv6 | Single-stage object detection framework dedicated to industrial applications | \u003Cul>\u003Cli>[Kaiheng Weng](https:\u002F\u002Fgithub.com\u002FkhwengXU)\u003C\u002Fli> \u003Cli>[Meng Cheng](https:\u002F\u002Fgithub.com\u002FMTChengMeng)\u003C\u002Fli> \u003Cli>[Yiduo Li](https:\u002F\u002Fgithub.com\u002Fyili123123)\u003C\u002Fli> \u003Cli>[Xiangxiang Chu](https:\u002F\u002Fscholar.google.com\u002Fcitations?&user=jn21pUsAAAAJ)\u003C\u002Fli> \u003Cli>[Xiaolin Wei](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=s5b7lU4AAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmeituan\u002FYOLOv6?style=social)](https:\u002F\u002Fgithub.com\u002Fmeituan\u002FYOLOv6) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.02976), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.05586)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Flearnopencv.com\u002Fyolov6-object-detection\u002F)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcocodataset.org\u002F#download)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fyolov6-docs.readthedocs.io\u002Fzh_CN\u002Flatest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFeiGeChuanShu\u002Fncnn-android-yolov6), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDefTruth\u002Flite.ai.toolkit\u002Fblob\u002Fmain\u002Flite\u002Fort\u002Fcv\u002Fyolov6.cpp), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FLinaom1214\u002FTensorRT-For-YOLO-Series), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhiqwang\u002Fyolov5-rt-stack\u002Ftree\u002Fmain\u002Fdeployment\u002Ftensorrt-yolov6)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3OpwcGU7VvE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGJ0lVOE3a7c), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3hqkbqJ5ag8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfFCWrMFH2UY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmeituan\u002FYOLOv6\u002Fblob\u002Fmaster\u002Fturtorial.ipynb) | 08.10.2023 |\n| DreamGaussian | Algorithm to convert 3D Gaussians into textured meshes and apply a fine-tuning stage to refine the details | \u003Cul>\u003Cli>[Jiaxiang Tang](https:\u002F\u002Fme.kiui.moe\u002F)\u003C\u002Fli> \u003Cli>[Jiawei Ren](https:\u002F\u002Fjiawei-ren.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hang Zhou](https:\u002F\u002Fhangz-nju-cuhk.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gang Zeng](http:\u002F\u002Fwww.cis.pku.edu.cn\u002Finfo\u002F1177\u002F1378.htm)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdreamgaussian\u002Fdreamgaussian?style=social)](https:\u002F\u002Fgithub.com\u002Fdreamgaussian\u002Fdreamgaussian) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.16653)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgraphdeco-inria\u002Fdiff-gaussian-rasterization), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fnvdiffrast), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhoffstadt\u002FDearPyGui)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fdreamgaussian.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1sLpYmmLS209-e5eHgcuqdryFRRO6ZhFS) | 04.10.2023 |\n| ICON | Given a set of images, method estimates a detailed 3D surface from each image and then combines these into an animatable avatar | \u003Cul>\u003Cli>[Yuliang Xiu](https:\u002F\u002Fxiuyuliang.cn\u002F)\u003C\u002Fli> \u003Cli>[Jinlong Yang](https:\u002F\u002Fis.mpg.de\u002F~jyang)\u003C\u002Fli> \u003Cli>[Dimitrios Tzionas](https:\u002F\u002Fps.is.mpg.de\u002F~dtzionas)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2bcf751092ab.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01294) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyuliangxiu\u002Ficon?style=social)](https:\u002F\u002Fgithub.com\u002Fyuliangxiu\u002Ficon) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.09127)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FKeypointNeRF), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYadiraF\u002FPIXIE), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYuliangXiu\u002Fbvh-distance-queries), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FProject-Splinter\u002FMonoPortDataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZhengZerong\u002FPaMIR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FProject-Splinter\u002FMonoPort), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fshunsukesaito\u002FSCANimate), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Faistplusplus_api)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FYuliang\u002FICON)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ficon.is.tue.mpg.de\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhZd6AYin2DE)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1-AWeWhPvCTBX0KfMtgtMk10uPU05ihoA) | 31.08.2023 |\n| DINOv2 | Produce high-performance visual features that can be directly employed with classifiers as simple as linear layers on a variety of computer vision tasks; these visual features are robust and perform well across domains without any requirement for fine-tuning | \u003Cul>\u003Cli>[Maxime Oquab](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5vteYV8AAAAJ)\u003C\u002Fli> \u003Cli>[Timothée Darcet](https:\u002F\u002Fgithub.com\u002FTimDarcet)\u003C\u002Fli> \u003Cli>[Théo Moutakanni](https:\u002F\u002Fgithub.com\u002FTheoMoutakanni)\u003C\u002Fli> \u003Cli>[Huy Vo](https:\u002F\u002Fhuyvvo.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Marc Szafraniec](https:\u002F\u002Fgithub.com\u002FMarcSzafraniec\u002F)\u003C\u002Fli> \u003Cli>[Vasil Khalidov](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=tjazz3AAAAAJ)\u003C\u002Fli> \u003Cli>[Pierre Fernandez](https:\u002F\u002Fpierrefdz.github.io\u002F)\u003C\u002Fli> \u003Cli>[Daniel Haziza](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2eSKdFMAAAAJ)\u003C\u002Fli> \u003Cli>[Francisco Massa](https:\u002F\u002Fgithub.com\u002Ffmassa)\u003C\u002Fli> \u003Cli>[Alaaeldin El-Nouby](https:\u002F\u002Faelnouby.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mahmoud Assran](http:\u002F\u002Fwww.midoassran.ca\u002F)\u003C\u002Fli> \u003Cli>[Nicolas Ballas](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=euUV4iUAAAAJ)\u003C\u002Fli> \u003Cli>[Wojciech Galuba](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jyaTX64AAAAJ)\u003C\u002Fli> \u003Cli>[Russell Howes](http:\u002F\u002Fwww.russellhowes.net\u002F)\u003C\u002Fli> \u003Cli>[Po-Yao Huang](https:\u002F\u002Fberniebear.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shang-Wen Li](https:\u002F\u002Fswdanielli.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ishan Misra](http:\u002F\u002Fimisra.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michael Rabbat](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=cMPKe9UAAAAJ)\u003C\u002Fli> \u003Cli>[Vasu Sharma](https:\u002F\u002Fvasusharma.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gabriel Synnaeve](https:\u002F\u002Fsyhw.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hu Xu](https:\u002F\u002Fhowardhsu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hervé Jégou](https:\u002F\u002Fgithub.com\u002Fjegou)\u003C\u002Fli> \u003Cli>[Julien Mairal](http:\u002F\u002Fthoth.inrialpes.fr\u002Fpeople\u002Fmairal\u002F)\u003C\u002Fli> \u003Cli>[Patrick Labatut](https:\u002F\u002Fgithub.com\u002Fpatricklabatut)\u003C\u002Fli> \u003Cli>[Armand Joulin](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=kRJkDakAAAAJ)\u003C\u002Fli> \u003Cli>[Piotr Bojanowski](https:\u002F\u002Fgithub.com\u002Fpiotr-bojanowski)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fdinov2?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdinov2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.07193)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fai.facebook.com\u002Fblog\u002Fdino-v2-computer-vision-self-supervised-learning\u002F)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fdinov2.metademolab.com\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmain\u002Fmodel_doc\u002Fdinov2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fpurnasaigudikandula.medium.com\u002Fdinov2-image-classification-visualization-and-paper-review-745bee52c826), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Ftowardsdatascience.com\u002Fmeta-ais-another-revolutionary-large-scale-model-dinov2-for-image-feature-extraction-1114b287eadd)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FcsEgtSh7jV4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FKSZiJ4k28b4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRZEkdOc3szU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fdinov2\u002Fblob\u002Fmain\u002Fnotebooks\u002Fsemantic_segmentation.ipynb) | 31.08.2023 |\n| StyleGAN 3 | Alias-Free Generative Adversarial Networks | \u003Cul>\u003Cli>[Tero Karras](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Ftero-karras)\u003C\u002Fli> \u003Cli>[Miika Aittala](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002FMiika-Aittala)\u003C\u002Fli> \u003Cli>[Samuli Laine](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002FSamuli-Laine)\u003C\u002Fli> \u003Cli>[Erik Härkönen](https:\u002F\u002Fgithub.com\u002Fharskish)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Janne Hellsten](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002FJanne-Hellsten)\u003C\u002Fli> \u003Cli>[Jaakko Lehtinen](https:\u002F\u002Fusers.aalto.fi\u002F~lehtinj7\u002F)\u003C\u002Fli> \u003Cli>[Timo Aila](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Ftimo-aila)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNVlabs\u002Fstylegan3?style=social)](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan3) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.12423), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.08500), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.01401), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.06991), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.04948), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1606.03498)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan3-detector), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fmetfaces-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2021\u002Fhash\u002F076ccd93ad68be51f23707988e934906-Abstract.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnvlabs.github.io\u002Fstylegan3)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1BXNHZBai-pXtP-ncliouXo_kUiG1Pq7M) | 13.08.2023 |\n| FateZero | Zero-shot text-based editing method on real-world videos without per-prompt training or use-specific mask | \u003Cul>\u003Cli>[Chenyang Qi](https:\u002F\u002Fchenyangqiqi.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002Facademic\u002F)\u003C\u002Fli> \u003Cli>[Yong Zhang](https:\u002F\u002Fyzhang2016.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chenyang Lei](https:\u002F\u002Fchenyanglei.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli> \u003Cli>[Qifeng Chen](https:\u002F\u002Fcqf.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a4c0093cdc32.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.01460) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FChenyangQiQi\u002FFateZero?style=social)](https:\u002F\u002Fgithub.com\u002FChenyangQiQi\u002FFateZero) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.09535)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbryandlee\u002FTune-A-Video), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fprompt-to-prompt)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fchenyangqi\u002FFateZero), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fchenyangqi\u002Fjeep_tuned_200)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ffate-zero-edit.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002F11uzioo\u002Fr_fatezero_fusing_attentions_for_zeroshot\u002F)\u003C\u002Fli>\u003Cli>[video](https:\u002F\u002Fhkustconnect-my.sharepoint.com\u002Fpersonal\u002Fcqiaa_connect_ust_hk\u002F_layouts\u002F15\u002Fstream.aspx?id=%2Fpersonal%2Fcqiaa%5Fconnect%5Fust%5Fhk%2FDocuments%2Fdiffusion%2Fweb%5Fvideo%2Emp4&ga=1&referrer=StreamWebApp%2EWeb&referrerScenario=AddressBarCopied%2Eview%2E9b85614a%2D5af9%2D4485%2Dbcb1%2Db39f90e8d381)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FChenyangQiQi\u002FFateZero\u002Fblob\u002Fmain\u002Fcolab_fatezero.ipynb) | 13.08.2023 |\n| Big GAN | Large Scale GAN Training for High Fidelity Natural Image Synthesis | \u003Cul>\u003Cli>[Andrew Brock](https:\u002F\u002Fgithub.com\u002Fajbrock)\u003C\u002Fli> \u003Cli>[Jeff Donahue](https:\u002F\u002Fjeffdonahue.com\u002F)\u003C\u002Fli> \u003Cli>[Karen Simonyan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=L7lMQkQAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.11096)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fhub\u002Fblob\u002Fmaster\u002Fexamples\u002Fcolab\u002Fbiggan_generation_with_tf_hub.ipynb) | 03.08.2023 |\n| LaMa | Resolution-robust Large Mask Inpainting with Fourier Convolutions | \u003Cul>\u003Cli>[Roman Suvorov](https:\u002F\u002Fgithub.com\u002Fwindj007)\u003C\u002Fli> \u003Cli>[Elizaveta Logacheva](https:\u002F\u002Fgithub.com\u002Felimohl)\u003C\u002Fli> \u003Cli>[Anton Mashikhin](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fheyt0ny\u002F)\u003C\u002Fli> \u003Cli>[Anastasia Remizova](https:\u002F\u002Fgithub.com\u002Ffeathernox)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Arsenii Ashukha](https:\u002F\u002Fashukha.com\u002F)\u003C\u002Fli> \u003Cli>[Aleksei Silvestrov](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002F%D0%B0%D0%BB%D0%B5%D0%BA%D1%81%D0%B5%D0%B9-%D1%81%D0%B8%D0%BB%D1%8C%D0%B2%D0%B5%D1%81%D1%82%D1%80%D0%BE%D0%B2-141b99b6\u002F)\u003C\u002Fli> \u003Cli>[Naejin Kong](https:\u002F\u002Fgithub.com\u002Fnaejin-kong)\u003C\u002Fli> \u003Cli>[Harshith Goka](https:\u002F\u002Fgithub.com\u002Fh9399-goka)\u003C\u002Fli> \u003Cli>[Kiwoong Park](https:\u002F\u002Fgithub.com\u002Fkyoong-park)\u003C\u002Fli> \u003Cli>[Victor Lempitsky](http:\u002F\u002Fsites.skoltech.ru\u002Fcompvision\u002Fmembers\u002Fvilem\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_f3f60f020173.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FWACV51458.2022.00323) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsaic-mdal\u002Flama?style=social)](https:\u002F\u002Fgithub.com\u002Fsaic-mdal\u002Flama) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.07161)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fandy971022\u002Fauto-lama), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frichzhang\u002FPerceptualSimilarity), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FPo-Hsun-Su\u002Fpytorch-ssim), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmseitzer\u002Fpytorch-fid)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsaic-mdal.github.io\u002Flama-project\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsaic-mdal\u002Flama\u002Fblob\u002Fmaster\u002Fcolab\u002FLaMa_inpainting.ipynb) | 02.08.2023 |\n| MakeItTalk | A method that generates expressive talking-head videos from a single facial image with audio as the only input | \u003Cul>\u003Cli>[Yang Zhou](https:\u002F\u002Fpeople.umass.edu\u002F~yangzhou\u002F)\u003C\u002Fli> \u003Cli>[Xintong Han](http:\u002F\u002Fusers.umiacs.umd.edu\u002F~xintong\u002F)\u003C\u002Fli> \u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli> \u003Cli>[Jose Echevarria](http:\u002F\u002Fwww.jiechevarria.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Evangelos Kalogerakis](https:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002F)\u003C\u002Fli> \u003Cli>[Dingzeyu Li](https:\u002F\u002Fdingzeyu.li\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_84498a2017f8.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3414685.3417774) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyzhou359\u002FMakeItTalk?style=social)](https:\u002F\u002Fgithub.com\u002Fyzhou359\u002FMakeItTalk) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.12992)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fdrive.google.com\u002Fdrive\u002Ffolders\u002F1EwuAy3j1b9Zc1MsidUfxG_pJGc_cV60O)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpeople.umass.edu\u002F~yangzhou\u002FMakeItTalk\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=vUMGKASgbf8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fiboyles\u002Fmakeittalknow\u002Fblob\u002Fmain\u002Fworking_quick_demo_of_makeittalk_07_2023.ipynb) | 27.07.2023 |\n| HiDT | A generative image-to-image model and a new upsampling scheme that allows to apply image translation at high resolution | \u003Cul>\u003Cli>[Denis Korzhenkov](https:\u002F\u002Fgithub.com\u002Fdenkorzh)\u003C\u002Fli> \u003Cli>[Gleb Sterkin](https:\u002F\u002Fgithub.com\u002Fbelkakari)\u003C\u002Fli> \u003Cli>[Sergey Nikolenko](https:\u002F\u002Flogic.pdmi.ras.ru\u002F~sergey\u002F)\u003C\u002Fli> \u003Cli>[Victor Lempitsky](http:\u002F\u002Fsites.skoltech.ru\u002Fcompvision\u002Fmembers\u002Fvilem\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_8388d5acf22d.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00751) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsaic-mdal\u002FHiDT?style=social)](https:\u002F\u002Fgithub.com\u002Fsaic-mdal\u002FHiDT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.08791)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsaic-mdal.github.io\u002FHiDT\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLuvGzlEQXT1KQuKrfBBEWh2f3PToxyeM5), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=EWKAgwgqXB4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsaic-mdal\u002Fhidt\u002Fblob\u002Fmaster\u002Fnotebooks\u002FHighResolutionDaytimeTranslation.ipynb) | 24.07.2023 |\n| AWQ | Activation-aware Weight Quantization, a hardware-friendly approach for LLM low-bit weight-only quantization | \u003Cul>\u003Cli>[Ji Lin](http:\u002F\u002Flinji.me\u002F)\u003C\u002Fli> \u003Cli>[Jiaming Tang](http:\u002F\u002Fjiamingtang.me\u002F)\u003C\u002Fli> \u003Cli>[Haotian Tang](https:\u002F\u002Fkentang.net\u002F)\u003C\u002Fli> \u003Cli>[Shang Yang](https:\u002F\u002Fgithub.com\u002Fys-2020)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Wei-Ming Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=6xFvyJwAAAAJ)\u003C\u002Fli> \u003Cli>[Wei-Chen Wang](https:\u002F\u002Fweichenwang.me\u002F)\u003C\u002Fli> \u003Cli>[Guangxuan Xiao](https:\u002F\u002Fguangxuanx.com\u002F)\u003C\u002Fli> \u003Cli>[Xingyu Dang](https:\u002F\u002Fgithub.com\u002Fdangxingyu)\u003C\u002Fli> \u003Cli>[Chuang Gan](https:\u002F\u002Fgithub.com\u002Fchuangg)\u003C\u002Fli> \u003Cli>[Song Han](https:\u002F\u002Fsonghan.mit.edu\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_260e9c4ff296.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3714983.3714987) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmit-han-lab\u002Fllm-awq?style=social)](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fllm-awq) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.00978), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.07533), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.17323)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fvila.hanlab.ai\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvllm-project\u002Fvllm\u002Fblob\u002Fmain\u002Fvllm\u002Fmodel_executor\u002Flayers\u002Fquantization\u002Fawq.py), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FTensorRT-LLM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Ftext-generation-inference\u002Fpull\u002F1054), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcasper-hansen\u002FAutoAWQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002FVILA), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fsmoothquant)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fmit-han-lab\u002Fawq-model-zoo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fbyte-sized-ai\u002Fvllm-quantization-awq-activation-aware-weight-quantization-for-llm-compression-and-35894ffd6a9b)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fhanlab.mit.edu\u002Fprojects\u002Fawq)\u003C\u002Fli>\u003Cli>[slides](https:\u002F\u002Fwww.dropbox.com\u002Fscl\u002Ffi\u002Fdtnp6h6y1mnp7g036axu6\u002FAWQ-slide.pdf)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdcINVsqxQgQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3dYLj9vjfA0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FOMkyocVyEpQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmit-han-lab\u002Fllm-awq\u002Fblob\u002Fmaster\u002Fexamples\u002Fchat_demo.ipynb) | 24.07.2023 |\n| Once-for-All | Train a once-for-all network that supports diverse architectural settings by decoupling training and search, to reduce the cost | \u003Cul>\u003Cli>[Han Cai](https:\u002F\u002Fhan-cai.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chuang Gan](https:\u002F\u002Fpeople.csail.mit.edu\u002Fganchuang\u002F)\u003C\u002Fli> \u003Cli>[Tianzhe Wang](https:\u002F\u002Fsites.google.com\u002Fview\u002Ftianzhe-wang\u002Fhome)\u003C\u002Fli> \u003Cli>[Zhekai Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=aYh01A4AAAA)\u003C\u002Fli> \u003Cli>[Song Han](https:\u002F\u002Fhanlab.mit.edu\u002Fsonghan)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_3ec06e4edff8.png)](https:\u002F\u002Fdoi.org\u002F10.48550\u002FarXiv.1908.09791) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmit-han-lab\u002Fonce-for-all?style=social)](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fonce-for-all) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.00332), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.03494), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.08886)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsony\u002Fnnabla-nas), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fanalogdevicesinc\u002Fai8x-training)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fpytorch.org\u002Fhub\u002Fpytorch_vision_once_for_all)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fofa\u002F)\u003C\u002Fli>\u003Cli>[slides](https:\u002F\u002Ffile.lzhu.me\u002Fprojects\u002FOnceForAll\u002FOFA%20Slides.pdf)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fa_OeT8MXzWI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fwrsid5tvuSM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfptQ_eJ3Uc0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjsyHqDX5cU8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fg3ujV7q0wZk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FLgt2kg1_pTI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FFVfh2vw4RG0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FWHQWlIKdwsk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmit-han-lab\u002Fonce-for-all\u002Fblob\u002Fmaster\u002Ftutorial\u002Fofa.ipynb) | 19.07.2023 |\n| Recognize Anything & Tag2Text | Vision language pre-training framework, which introduces image tagging into vision-language models to guide the learning of visual-linguistic features | \u003Cul>\u003Cli>[Xinyu Huang](https:\u002F\u002Fxinyu1205.github.io\u002F)\u003C\u002Fli> \u003Cli>[Youcai Zhang](https:\u002F\u002Fgithub.com\u002FColer1994)\u003C\u002Fli> \u003Cli>[Jinyu Ma](https:\u002F\u002Fgithub.com\u002Fmajinyu666)\u003C\u002Fli> \u003Cli>[Zhaoyang Li](https:\u002F\u002Fgithub.com\u002FZhaoyangLi-nju)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yanchun Xie](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=T0xk9-wAAAAJ)\u003C\u002Fli> \u003Cli>[Yuzhuo Qin](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5ZG65AkAAAAJ)\u003C\u002Fli> \u003Cli>[Tong Luo](https:\u002F\u002Fieeexplore.ieee.org\u002Fauthor\u002F37089387319)\u003C\u002Fli> \u003Cli>[Yaqian Li](https:\u002F\u002Fopenreview.net\u002Fprofile?id=~Yaqian_Li1)\u003C\u002Fli> \u003Cli>[Yandong Guo](http:\u002F\u002Fwww.lsl.zone\u002F)\u003C\u002Fli> \u003Cli>[Yandong Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=fWDoWsQAAAAJ)\u003C\u002Fli> \u003Cli>[Lei Zhang](https:\u002F\u002Fwww.leizhang.org\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fxinyu1205\u002Frecognize-anything?style=social)](https:\u002F\u002Fgithub.com\u002Fxinyu1205\u002Frecognize-anything) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.03514), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.05657)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOpenGVLab\u002FAsk-Anything), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fpositive666\u002FPrompt-Can-Anything)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-recognize-anything-a-strong-image-tagging-model-9e5e1c6dd0af)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Frecognize-anything.github.io\u002F), [project](https:\u002F\u002Frecognize-anything.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmhd-medfa\u002Frecognize-anything\u002Fblob\u002Fmain\u002Frecognize_anything_demo.ipynb) | 09.07.2023 |\n| Thin-Plate Spline Motion Model | End-to-end unsupervised motion transfer framework | \u003Cul>\u003Cli>[Jian Zhao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=OKm5CQYAAAAJ)\u003C\u002Fli> \u003Cli>[Hui Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=w3mzCiwAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_60204a493e21.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00364) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyoyo-nb\u002FThin-Plate-Spline-Motion-Model?style=social)](https:\u002F\u002Fgithub.com\u002Fyoyo-nb\u002FThin-Plate-Spline-Motion-Model) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.14367)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Fmonkey-net), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Fvideo-preprocessing), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Fpose-evaluation), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTalkUHulk\u002FImage-Animation-Turbo-Boost)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FCVPR\u002FImage-Animation-using-Thin-Plate-Spline-Motion-Model)\u003C\u002Fli>\u003Cli>[supp](https:\u002F\u002Fcloud.tsinghua.edu.cn\u002Ff\u002Ff7b8573bb5b04583949f\u002F?dl=1)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1DREfdpnaBhqISg0fuQlAAIwyGVn1loH_) | 07.07.2023 |\n| MobileSAM | Towards Lightweight SAM for Mobile Applications | \u003Cul>\u003Cli>[Chaoning Zhang](https:\u002F\u002Fgithub.com\u002FChaoningZhang)\u003C\u002Fli> \u003Cli>[Dongshen Han](https:\u002F\u002Fgithub.com\u002Fdongshenhan)\u003C\u002Fli> \u003Cli>[Yu Qiao](https:\u002F\u002Fgithub.com\u002Fqiaoyu1002)\u003C\u002Fli> \u003Cli>[Jung Uk Kim](https:\u002F\u002Fvisualai.khu.ac.kr\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Sung-Ho Bae](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=EULut5oAAAAJ)\u003C\u002Fli> \u003Cli>[Seungkyu Lee](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=3Pf6C6cAAAAJ)\u003C\u002Fli> \u003Cli>[Choong Seon Hong](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=oKANWloAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FChaoningZhang\u002FMobileSAM?style=social)](https:\u002F\u002Fgithub.com\u002FChaoningZhang\u002FMobileSAM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.14289)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjolibrain\u002FjoliGEN), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fakbartus\u002FMobileSAM-in-the-Browser), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fqiaoyu1002\u002FInpaint-Anything), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fqiaoyu1002\u002FPersonalize-SAM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJumpat\u002FSegmentAnythingin3D), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvietanhdev\u002Fanylabeling), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fwangsssky\u002FSonarSAM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcontinue-revolution\u002Fsd-webui-segment-anything)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002F_akhaliq\u002Fstatus\u002F1674410573075718145)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FeTEfq_kWabQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FChaoningZhang\u002FMobileSAM\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fpredictor_example.ipynb) | 30.06.2023 |\n| T5X | Modular, composable, research-friendly framework for high-performance, configurable, self-service training, evaluation, and inference of sequence models at many scales | \u003Cul>\u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Hyung Won Chung](https:\u002F\u002Fgithub.com\u002Fhwchung27)\u003C\u002Fli> \u003Cli>[Anselm Levskaya](https:\u002F\u002Fanselmlevskaya.com\u002F)\u003C\u002Fli> \u003Cli>[Gaurav Mishra](https:\u002F\u002Fresearch.google\u002Fpeople\u002FGauravMishra\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[James Bradbury](https:\u002F\u002Fgithub.com\u002Fjekbradbury)\u003C\u002Fli> \u003Cli>[Daniel Andor](https:\u002F\u002Fgithub.com\u002Fandorardo)\u003C\u002Fli> \u003Cli>[Sharan Narang](https:\u002F\u002Fgithub.com\u002Fsharannarang)\u003C\u002Fli> \u003Cli>[Brian Lester](https:\u002F\u002Fblester125.com\u002F)\u003C\u002Fli> \u003Cli>[Colin Gaffney](https:\u002F\u002Fgithub.com\u002Fcpgaffney1)\u003C\u002Fli> \u003Cli>[Afroz Mohiuddin](https:\u002F\u002Fgithub.com\u002Fafrozenator)\u003C\u002Fli> \u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli> \u003Cli>[Aitor Lewkowycz](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Yum1ah0AAAAJ)\u003C\u002Fli> \u003Cli>[Alex Salcianu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=HSrT1wsAAAAJ)\u003C\u002Fli> \u003Cli>[Marc van Zee](https:\u002F\u002Fgithub.com\u002Fmarcvanzee)\u003C\u002Fli> \u003Cli>[Jacob Austin](https:\u002F\u002Fjacobaustin123.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sebastian Goodman](https:\u002F\u002Fgithub.com\u002F0x0539)\u003C\u002Fli> \u003Cli>[Livio Baldini Soares](https:\u002F\u002Fliviosoares.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haitang Hu](https:\u002F\u002Fhthu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sasha Tsvyashchenko](https:\u002F\u002Fendl.ch\u002F)\u003C\u002Fli> \u003Cli>[Aakanksha Chowdhery](http:\u002F\u002Fwww.achowdhery.com\u002F)\u003C\u002Fli> \u003Cli>[Jasmijn Bastings](https:\u002F\u002Fjasmijn.ninja\u002F)\u003C\u002Fli> \u003Cli>[Jannis Bulian](http:\u002F\u002Fbulian.org\u002F)\u003C\u002Fli> \u003Cli>[Xavier Garcia](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Y2Hio6MAAAAJ)\u003C\u002Fli> \u003Cli>[Jianmo Ni](https:\u002F\u002Fnijianmo.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kathleen Kenealy](https:\u002F\u002Fscholar.google.com\u002Fcitations?&user=HgRBC5gAAAAJ)\u003C\u002Fli> \u003Cli>[Jonathan Clark](http:\u002F\u002Fwww.cs.cmu.edu\u002F~jhclark\u002F)\u003C\u002Fli> \u003Cli>[Dan Garrette](http:\u002F\u002Fwww.dhgarrette.com\u002F)\u003C\u002Fli> \u003Cli>[James Lee-Thorp](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=qsPv098AAAAJ)\u003C\u002Fli> \u003Cli>[Colin Raffel](https:\u002F\u002Fcolinraffel.com\u002F)\u003C\u002Fli> \u003Cli>[Noam Shazeer](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=wsGvgA8AAAAJ)\u003C\u002Fli> \u003Cli>[Marvin Ritter](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=arcf5FgAAAAJ)\u003C\u002Fli> \u003Cli>[Maarten Bosma](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=wkeFQPgAAAAJ)\u003C\u002Fli> \u003Cli>[Alexandre Passos](https:\u002F\u002Fwww.ic.unicamp.br\u002F~tachard\u002F)\u003C\u002Fli> \u003Cli>[Jeremy Maitin-Shepard](https:\u002F\u002Fresearch.google\u002Fpeople\u002FJeremyMaitinShepard\u002F)\u003C\u002Fli> \u003Cli>[Noah Fiedel](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=XWpV9DsAAAAJ)\u003C\u002Fli> \u003Cli>[Brennan Saeta](https:\u002F\u002Fgithub.com\u002Fsaeta)\u003C\u002Fli> \u003Cli>[Ryan Sepassi](https:\u002F\u002Fryansepassi.com\u002F)\u003C\u002Fli> \u003Cli>[Alexander Spiridonov](https:\u002F\u002Fresearch.google\u002Fpeople\u002FAlexanderSpiridonov\u002F)\u003C\u002Fli> \u003Cli>[Joshua Newlan](https:\u002F\u002Fgithub.com\u002Fjoshnewlan)\u003C\u002Fli> \u003Cli>[Andrea Gesmundo](https:\u002F\u002Fgithub.com\u002Fagesmundo)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Ft5x?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Ft5x) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.17189), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.10683)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Ft5x.readthedocs.io\u002Fen\u002Flatest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fmesh), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fserving)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fdatasets\u002Fcatalog\u002Fwmt_t2t_translate), [\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fguide\u002Fdata), [\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Ftensorboard)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Ft5x\u002Fblob\u002Fmain\u002Ft5x\u002Fnotebooks\u002Fintroduction.ipynb) | 27.06.2023 |\n| CodeTalker | Cast speech-driven facial animation as a code query task in a finite proxy space of the learned codebook, which effectively promotes the vividness of the generated motions by reducing the cross-modal mapping uncertainty | \u003Cul>\u003Cli>[Jinbo Xing](https:\u002F\u002Fdoubiiu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Menghan Xia](https:\u002F\u002Fmenghanxia.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuechen Zhang](https:\u002F\u002Fjulianjuaner.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002Facademic\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jue Wang](https:\u002F\u002Fjuewang725.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tien-Tsin Wong](https:\u002F\u002Fttwong12.github.io\u002Fmyself.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_de3910b4cb3f.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52729.2023.01229) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FDoubiiu\u002FCodeTalker?style=social)](https:\u002F\u002Fgithub.com\u002FDoubiiu\u002FCodeTalker) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.02379), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.09797)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMPI-IS\u002Fmesh), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTimoBolkart\u002Fvoca\u002Ftree\u002Fmaster\u002Ftemplate), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEvelynFan\u002FFaceFormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FRenYurui\u002FPIRender), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOpenTalker\u002FStyleHEAT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMeta-Portrait\u002FMetaPortrait)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fdoubiiu.github.io\u002Fprojects\u002Fcodetalker\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJ2RngmuYrG4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FDoubiiu\u002FCodeTalker\u002Fblob\u002Fmain\u002Fdemo.ipynb) | 16.06.2023 |\n| Gen-L-Video | Extending off-the-shelf short video diffusion models for generating and editing videos comprising hundreds of frames with diverse semantic segments without introducing additional training, all while preserving content consistency | \u003Cul>\u003Cli>[Fu-Yun Wang](https:\u002F\u002Fg-u-n.github.io\u002F)\u003C\u002Fli> \u003Cli>[Wenshuo Chen](https:\u002F\u002Fgithub.com\u002Fwinshot-thu)\u003C\u002Fli> \u003Cli>[Guanglu Song](https:\u002F\u002Fsongguanglu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Han-Jia Ye](https:\u002F\u002Fgithub.com\u002FHan-Jia)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yu Liu](https:\u002F\u002Fliuyu.us\u002F)\u003C\u002Fli> \u003Cli>[Hongsheng Li](https:\u002F\u002Fwww.ee.cuhk.edu.hk\u002F~hsli\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FG-U-N\u002FGen-L-Video?style=social)](https:\u002F\u002Fgithub.com\u002FG-U-N\u002FGen-L-Video) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.18264)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flllyasviel\u002FControlNet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTencentARC\u002FT2I-Adapter), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsail-sg\u002FEditAnything)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fg-u-n.github.io\u002Fprojects\u002Fgen-long-video\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FG-U-N\u002FGen-L-Video\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fglv_colab.ipynb) | 04.06.2023 |\n| First Order Motion Model for Image Animation | Transferring facial movements from video to image | [Aliaksandr Siarohin](https:\u002F\u002Faliaksandrsiarohin.github.io\u002Faliaksandr-siarohin-website\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAliaksandrSiarohin\u002Ffirst-order-model?style=social)](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Ffirst-order-model) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F2019\u002Fhash\u002F31c0b36aef265d9221af80872ceb62f9-Abstract.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Faliaksandrsiarohin.github.io\u002Ffirst-order-model-website\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=u-0cQ-grXBQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAliaksandrSiarohin\u002Ffirst-order-model\u002Fblob\u002Fmaster\u002Fdemo.ipynb) | 04.06.2023 |\n| PolyGen | Approach which models the mesh directly, predicting mesh vertices and faces sequentially using a Transformer-based architecture | \u003Cul>\u003Cli>[Charlie Nash](https:\u002F\u002Fgithub.com\u002Fcharlienash)\u003C\u002Fli> \u003Cli>[Yaroslav Ganin](https:\u002F\u002Fyaroslav.ganin.net\u002F)\u003C\u002Fli> \u003Cli>[Ali Eslami](http:\u002F\u002Farkitus.com\u002F)\u003C\u002Fli> \u003Cli>[Peter Battaglia](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=nQ7Ij30AAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Fdeepmind-research?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fdeepmind-research\u002Ftree\u002Fmaster\u002Fpolygen) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2002.10880), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1506.03134), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.04887)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fanshulcgm\u002Fpolygen)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FXCrjpIRkVCU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Fdeepmind-research\u002Fblob\u002Fmaster\u002Fpolygen\u002Ftraining.ipynbv) | 02.06.2023 |\n| Parallel WaveGAN | State-of-the-art non-autoregressive models to build your own great vocoder | [Tomoki Hayashi](https:\u002F\u002Fkan-bayashi.github.io\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_137784a250dd.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICASSP40776.2020.9053795) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fkan-bayashi\u002FParallelWaveGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fkan-bayashi\u002FParallelWaveGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.11480), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.06711), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.05106)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fkan-bayashi.github.io\u002FParallelWaveGAN\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Ftacotron2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fespnet\u002Fespnet)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fespnet\u002Fnotebook\u002Fblob\u002Fmaster\u002Fespnet2_tts_realtime_demo.ipynb) | 01.06.2023 |\n| ECON | designed for \"Human digitization from a color image\", which combines the best properties of implicit and explicit representations, to infer high-fidelity 3D clothed humans from in-the-wild images, even with loose clothing or in challenging poses | \u003Cul>\u003Cli>[Yuliang Xiu](https:\u002F\u002Fxiuyuliang.cn\u002F)\u003C\u002Fli> \u003Cli>[Jinlong Yang](https:\u002F\u002Fis.mpg.de\u002F~jyang)\u003C\u002Fli> \u003Cli>[Xu Cao](https:\u002F\u002Fxucao-42.github.io\u002Fhomepage\u002F)\u003C\u002Fli> \u003Cli>[Dimitrios Tzionas](https:\u002F\u002Fps.is.mpg.de\u002F~dtzionas)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ead37f141dbf.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52729.2023.00057) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYuliangXiu\u002FECON?style=social)](https:\u002F\u002Fgithub.com\u002FYuliangXiu\u002FECON) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.07422)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FVqa7KBGRyk)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYuliangXiu\u002FECON\u002Fblob\u002Fmaster\u002Fdocs\u002Finstallation-docker.md)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkwan3854\u002FCEB_ECON), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxucao-42\u002Fbilateral_normal_integration), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FProject-Splinter\u002FMonoPortDataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuangyangyi\u002FTeCH), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuangyangyi\u002FTeCH), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyfeng95\u002FPIXIE)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1451sjr\u002Fecon_explicit_clothed_humans_optimized_via_normal\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002Fyuliangxiu)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FsbWZbTf6ZYk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FSDVfCeaI4AY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5PEd_p90kS0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FMDFvV7y5Qgk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1YRgwoRCZIrSB2e7auEWFyG10Xzjbrbno) | 31.05.2023 |\n| MMS | The Massively Multilingual Speech project expands speech technology from about 100 languages to over 1000 by building a single multilingual speech recognition model supporting over 1100 languages, language identification models able to identify over 4000 languages, pretrained models supporting over 1400 languages, and text-to-speech models for over 1100 languages | \u003Cul>\u003Cli>[Vineel Pratap](https:\u002F\u002Fgithub.com\u002Fvineelpratap)\u003C\u002Fli> \u003Cli>[Andros Tjandra](https:\u002F\u002Fgithub.com\u002Fandrostj)\u003C\u002Fli> \u003Cli>[Bowen Shi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=xqyoorYAAAAJ)\u003C\u002Fli> \u003Cli>[Paden Tomasello](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=sBtWMGYAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Arun Babu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=oJfoTakAAAAJ)\u003C\u002Fli> \u003Cli>[Sayani Kundu](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsayani-kundu)\u003C\u002Fli> \u003Cli>[Ali Elkahky](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=KB3S8RoAAAAJ)\u003C\u002Fli> \u003Cli>[Zhaoheng Ni](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=SYFMSNsAAAAJ)\u003C\u002Fli> \u003Cli>[Apoorv Vyas](https:\u002F\u002Fapoorv2904.github.io\u002F)\u003C\u002Fli> \u003Cli>[Maryam Fazel-Zarandi](https:\u002F\u002Fwww.maryamfazel.com\u002F)\u003C\u002Fli> \u003Cli>[Alexei Baevski](https:\u002F\u002Fgithub.com\u002Falexeib)\u003C\u002Fli> \u003Cli>[Yossi Adi](https:\u002F\u002Fwww.cs.huji.ac.il\u002F~adiyoss\u002F)\u003C\u002Fli> \u003Cli>[Xiaohui Zhang](https:\u002F\u002Fgithub.com\u002Fxiaohui-zhang)\u003C\u002Fli> \u003Cli>[Wei-Ning Hsu](https:\u002F\u002Fwnhsu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Alexis Conneau](https:\u002F\u002Fgithub.com\u002Faconneau)\u003C\u002Fli> \u003Cli>[Michael Auli](https:\u002F\u002Fgithub.com\u002Fmichaelauli)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Ffairseq?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffairseq\u002Ftree\u002Fmain\u002Fexamples\u002Fmms) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.13516)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmain\u002Fen\u002Fmodel_doc\u002Fmms), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002Fmms-cclms\u002F), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fmms_adapters)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.facebook.com\u002Fblog\u002Fmultilingual-model-speech-recognition\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGEzxHxWys2s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fg06agCmxS7I)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Ffairseq\u002Fblob\u002Fmain\u002Fexamples\u002Fmms\u002Fasr\u002Ftutorial\u002FMMS_ASR_Inference_Colab.ipynb) | 26.05.2023 |\n| FAB | Flow AIS Bootstrap uses AIS to generate samples in regions where the flow is a poor approximation of the target, facilitating the discovery of new modes | \u003Cul>\u003Cli>[Laurence Midgley](https:\u002F\u002Flollcat.github.io\u002Flaurence-midgley\u002F)\u003C\u002Fli> \u003Cli>[Vincent Stimper](https:\u002F\u002Fis.mpg.de\u002Fperson\u002Fvstimper)\u003C\u002Fli> \u003Cli>[Gregor N. C. Simm](https:\u002F\u002Fwww.gncs.me\u002F)\u003C\u002Fli> \u003Cli>[Bernhard Schölkopf](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=DZ-fHPgAAAAJ)\u003C\u002Fli> \u003Cli>[José Miguel Hernández-Lobato](https:\u002F\u002Fjmhl.org\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Flollcat\u002Ffab-torch?style=social)](https:\u002F\u002Fgithub.com\u002Flollcat\u002Ffab-torch) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.01893)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flollcat\u002Ffab-jax-old), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fannealed_flow_transport)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FxQQXvOWu9nE)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Flollcat\u002Ffab-torch\u002Fblob\u002Fmaster\u002Fexperiments\u002Fgmm\u002Ffab_gmm.ipynb) | 29.04.2023 |\n| CodeFormer | Transformer-based prediction network to model global composition and context of the low-quality faces for code prediction, enabling the discovery of natural faces that closely approximate the target faces even when the inputs are severely degraded | \u003Cul>\u003Cli>[Shangchen Zhou](https:\u002F\u002Fshangchenzhou.com\u002F)\u003C\u002Fli> \u003Cli>[Kelvin Chan](https:\u002F\u002Fckkelvinchan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chongyi Li](https:\u002F\u002Fli-chongyi.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsczhou\u002FCodeFormer?style=social)](https:\u002F\u002Fgithub.com\u002Fsczhou\u002FCodeFormer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.11253)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsamb-t\u002Funleashing-transformers), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepcam-cn\u002Fyolov5-face), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002Ffacexlib)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002Fc573258c38d0a3919d8c1364053c45df-Abstract-Conference.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fshangchenzhou.com\u002Fprojects\u002FCodeFormer\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fd3VDpkXlueI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPtwWu-FugbA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FORtYP8NW4T0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fxc5lKOKBCcg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1m52PNveE4PBhYrecj34cnpEeiHcC5LTb) | 21.04.2023 |\n| Text2Video-Zero | Text-to-Image Diffusion Models are Zero-Shot Video Generators | \u003Cul>\u003Cli>[Levon Khachatryan](https:\u002F\u002Fgithub.com\u002Flev1khachatryan)\u003C\u002Fli> \u003Cli>[Andranik Movsisyan](https:\u002F\u002Fgithub.com\u002F19and99)\u003C\u002Fli> \u003Cli>[Vahram Tadevosyan](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fvtadevosian)\u003C\u002Fli> \u003Cli>[Roberto Henschel](https:\u002F\u002Fgithub.com\u002Frob-hen)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Zhangyang Wang](https:\u002F\u002Fwww.ece.utexas.edu\u002Fpeople\u002Ffaculty\u002Fatlas-wang)\u003C\u002Fli> \u003Cli>[Shant Navasardyan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=VJSh59sAAAAJ)\u003C\u002Fli> \u003Cli>[Humphrey Shi](https:\u002F\u002Fwww.humphreyshi.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_98f16e41baef.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.01462) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPicsart-AI-Research\u002FText2Video-Zero?style=social)](https:\u002F\u002Fgithub.com\u002FPicsart-AI-Research\u002FText2Video-Zero) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.13439), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.01341), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.17604)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdbolya\u002Ftomesd), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJiauZhang\u002FText2Video-Zero), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcamenduru\u002Ftext2video-zero-colab), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FSHI-Labs\u002FText2Video-Zero-sd-webui)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fdiffusers\u002Fapi\u002Fpipelines\u002Ftext_to_video_zero)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftext2video-zero.github.io\u002F)\u003C\u002Fli>\u003Cli>[video](https:\u002F\u002Fwww.dropbox.com\u002Fs\u002Fuv90mi2z598olsq\u002FText2Video-Zero.MP4)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbeeDJJz-Q0A), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F97-1GYPtz0M)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002Ftext2video-zero-colab\u002Fblob\u002Fmain\u002Ftext2video_all.ipynb) | 11.04.2023 |\n| Segment Anything | The Segment Anything Model produces high quality object masks from input prompts such as points or boxes, and it can be used to generate masks for all objects in an image | \u003Cul>\u003Cli>[Alexander Kirillov](https:\u002F\u002Falexander-kirillov.github.io\u002F)\u003C\u002Fli> \u003Cli>[Eric Mintun](https:\u002F\u002Fericmintun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nikhila Ravi](https:\u002F\u002Fnikhilaravi.com\u002F)\u003C\u002Fli> \u003Cli>[Hanzi Mao](https:\u002F\u002Fhanzimao.me\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chloé Rolland](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=n-SnMhoAAAAJ)\u003C\u002Fli> \u003Cli>[Laura Gustafson](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=c8IpF9gAAAAJ)\u003C\u002Fli> \u003Cli>[Tete Xiao](https:\u002F\u002Ftetexiao.com\u002F)\u003C\u002Fli> \u003Cli>[Spencer Whitehead](https:\u002F\u002Fwww.spencerwhitehead.com\u002F)\u003C\u002Fli> \u003Cli>[Alex Berg](http:\u002F\u002Facberg.com\u002F)\u003C\u002Fli> \u003Cli>[Wan-Yen Lo](https:\u002F\u002Fgithub.com\u002Fwanyenlo)\u003C\u002Fli> \u003Cli>[Piotr Dollár](https:\u002F\u002Fpdollar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ross Girshick](https:\u002F\u002Fwww.rossgirshick.info\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fsegment-anything?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fsegment-anything) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.02643)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fai.facebook.com\u002Fdatasets\u002Fsegment-anything\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.facebook.com\u002Fresearch\u002Fpublications\u002Fsegment-anything\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.facebook.com\u002Fblog\u002Fsegment-anything-foundation-model-image-segmentation\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fsegment-anything.com\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2O_vecl28OA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfVeW9a6wItM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FFjYE0tKWOiY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fsegment-anything\u002Fblob\u002Fmain\u002Fnotebooks\u002Fpredictor_example.ipynb) | 10.04.2023 |\n| FollowYourPose | Two-stage training scheme that can utilize image pose pair and pose-free video datasets and the pre-trained text-to-image model to obtain the pose-controllable character videos | \u003Cul>\u003Cli>[Yue Ma](https:\u002F\u002Fmayuelala.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yingqing He](https:\u002F\u002Fyingqinghe.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002Facademic\u002F)\u003C\u002Fli> \u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Siran Chen](https:\u002F\u002Fgithub.com\u002FSranc3)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli> \u003Cli>[Xiu Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Xrh1OIUAAAAJ)\u003C\u002Fli> \u003Cli>[Qifeng Chen](https:\u002F\u002Fcqf.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_452d0f81a9fa.png)](https:\u002F\u002Fdoi.org\u002F10.1609\u002Faaai.v38i5.28206) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmayuelala\u002FFollowYourPose?style=social)](https:\u002F\u002Fgithub.com\u002Fmayuelala\u002FFollowYourPose) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.01186), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10752)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbryandlee\u002FTune-A-Video)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FYueMafighting\u002FFollowYourPose_v1\u002Ftree\u002Fmain), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FCompVis\u002Fstable-diffusion-v1-4)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ffollow-your-pose.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmayuelala)\u003C\u002Fli>\u003Cli>[video](https:\u002F\u002Funderline.io\u002Flecture\u002F91712-follow-your-pose-pose-guided-text-to-video-generation-using-pose-free-videos)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmayuelala\u002FFollowYourPose\u002Fblob\u002Fmain\u002Fquick_demo.ipynb) | 07.04.2023 |\n| EVA3D | High-quality unconditional 3D human generative model that only requires 2D image collections for training | \u003Cul>\u003Cli>[Fangzhou Hong](https:\u002F\u002Fhongfz16.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhaoxi Chen](https:\u002F\u002Ffrozenburning.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yushi Lan](https:\u002F\u002Fgithub.com\u002FNIRVANALAN)\u003C\u002Fli> \u003Cli>[Liang Pan](https:\u002F\u002Fgithub.com\u002Fpaul007pl)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhongfz16\u002FEVA3D?style=social)](https:\u002F\u002Fgithub.com\u002Fhongfz16\u002FEVA3D) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.04888)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fhongfz16.github.io\u002Fprojects\u002FEVA3D.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJNV0FJ0aDWM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FM-kyvzTQrBI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fhongfz16\u002FEVA3D\u002Fblob\u002Fmain\u002Fnotebook\u002FEVA3D_Demo.ipynb) | 06.04.2023 |\n| Stable Dreamfusion | Using a pretrained 2D text-to-image diffusion model to perform text-to-3D synthesis | \u003Cul>\u003Cli>[Jiaxiang Tang](https:\u002F\u002Fme.kiui.moe\u002F)\u003C\u002Fli> \u003Cli>[Ben Poole](https:\u002F\u002Fcs.stanford.edu\u002F~poole\u002F)\u003C\u002Fli> \u003Cli>[Ajay Jain](https:\u002F\u002Fajayj.com\u002F)\u003C\u002Fli> \u003Cli>[Jon Barron](https:\u002F\u002Fjonbarron.info\u002F)\u003C\u002Fli> \u003Cli>[Ben Mildenhall](https:\u002F\u002Fbmild.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fashawkey\u002Fstable-dreamfusion?style=social)](https:\u002F\u002Fgithub.com\u002Fashawkey\u002Fstable-dreamfusion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.14988)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fashawkey\u002Ftorch-ngp), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhoffstadt\u002FDearPyGui)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Frunwayml\u002Fstable-diffusion-v1-5)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fdreamfusion3d.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fpytorch.org\u002Fdocs\u002Fstable\u002Fcpp_extension.html#torch.utils.cpp_extension.load)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuM5NPodZZ1U?t=219), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FzWD5ZR5GtJM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FL3G0dx1Q0R8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdIgDbBTztUM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1MXT3yfOFvO0ooKEfiUUvTKwUkrrlCHpF) | 04.04.2023 |\n| PIFuHD | Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization | \u003Cul>\u003Cli>[Shunsuke Saito](https:\u002F\u002Fshunsukesaito.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tomas Simon](http:\u002F\u002Fwww.cs.cmu.edu\u002F~tsimon\u002F)\u003C\u002Fli> \u003Cli>[Jason Saragih](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ss-IvjMAAAAJ)\u003C\u002Fli> \u003Cli>[Hanbyul Joo](https:\u002F\u002Fjhugestar.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_bb617c34b8dc.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00016) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fpifuhd?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fpifuhd) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.00452)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuEDqCxvF5yc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=8qnwbbDS8xk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F11z58bl3meSzo6kFqkahMa35G5jmh2Wgt) | 26.03.2023 |\n| VideoReTalking | System to edit the faces of a real-world talking head video according to input audio, producing a high-quality and lip-syncing output video even with a different emotion | \u003Cul>\u003Cli>[Kun Cheng](https:\u002F\u002Fgithub.com\u002Fkunncheng)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yong Zhang](https:\u002F\u002Fyzhang2016.github.io\u002F)\u003C\u002Fli> \u003Cli>[Menghan Xia](https:\u002F\u002Fmenghanxia.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Fei Yin](https:\u002F\u002Ffeiiyin.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mingrui Zhu](https:\u002F\u002Fweb.xidian.edu.cn\u002Fmrzhu\u002Fen\u002Findex.html)\u003C\u002Fli> \u003Cli>[Xuan Wang](https:\u002F\u002Fxuanwangvc.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jue Wang](https:\u002F\u002Fjuewang725.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nannan Wang](https:\u002F\u002Fweb.xidian.edu.cn\u002Fnnwang\u002Fen\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ff47e90dab4c.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3550469.3555399) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FOpenTalker\u002Fvideo-retalking?style=social)](https:\u002F\u002Fgithub.com\u002FOpenTalker\u002Fvideo-retalking) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.14758)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdonydchen\u002Fganimation_replicate), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FRenYurui\u002FPIRender), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOpenTalker\u002FStyleHEAT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFeiiYin\u002FSPI)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fxthemadgenius.medium.com\u002Fmaking-videos-talk-right-syncing-lips-with-sound-using-videoretalking-611428084bbc)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fopentalker.github.io\u002Fvideo-retalking\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F178krha\u002Fvideoretalking\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FpttsTrQ-fko), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2Lkw8AmmRn0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRJ8YK_K4Ne0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fvinthony\u002Fvideo-retalking\u002Fblob\u002Fmain\u002Fquick_demo.ipynb) | 19.03.2023 |\n| Visual ChatGPT | Connects ChatGPT and a series of Visual Foundation Models to enable sending and receiving images during chatting | \u003Cul>\u003Cli>[Chenfei Wu](https:\u002F\u002Fgithub.com\u002Fchenfei-wu)\u003C\u002Fli> \u003Cli>[Shengming Yin](https:\u002F\u002Fgithub.com\u002Fshengming-yin)\u003C\u002Fli> \u003Cli>[Weizhen Qi](https:\u002F\u002Fgithub.com\u002FWeizhenQ)\u003C\u002Fli> \u003Cli>[Xiaodong Wang](https:\u002F\u002Fwang-xiaodong1899.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Zecheng Tang](https:\u002F\u002Fgithub.com\u002FCODINNLG)\u003C\u002Fli> \u003Cli>[Nan Duan](https:\u002F\u002Fnanduan.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002Fvisual-chatgpt?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fvisual-chatgpt) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.04671)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhwchase17\u002Flangchain), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flllyasviel\u002FControlNet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftimothybrooks\u002Finstruct-pix2pix), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftimojl\u002Fclipseg)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0UfXlFUwLms), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F7YEiEyfPF5U)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F11BtP3h-w0dZjA-X8JsS9_eo8OeGYvxXB) | 15.03.2023 |\n| Tune-A-Video | One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation | \u003Cul>\u003Cli>[Jay Zhangjie Wu](https:\u002F\u002Fzhangjiewu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yixiao Ge](https:\u002F\u002Fgeyixiao.com\u002F)\u003C\u002Fli> \u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Stan Weixian Lei](https:\u002F\u002Fgithub.com\u002FStanLei52)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yuchao Gu](https:\u002F\u002Fycgu.site\u002F)\u003C\u002Fli> \u003Cli>[Yufei Shi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=rpnlkwEAAAAJ)\u003C\u002Fli> \u003Cli>[Wynne Hsu](https:\u002F\u002Fwww.comp.nus.edu.sg\u002F~whsu\u002F)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli> \u003Cli>[Xiaohu Qie](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=mk-F69UAAAAJ)\u003C\u002Fli> \u003Cli>[Mike Zheng Shou](https:\u002F\u002Fsites.google.com\u002Fview\u002Fshowlab)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_40d4b24cbec4.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.00701) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fshowlab\u002FTune-A-Video?style=social)](https:\u002F\u002Fgithub.com\u002Fshowlab\u002FTune-A-Video) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.11565), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10752)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FTune-A-Video-library), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fstabilityai\u002Fstable-diffusion-2-1), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fsd-dreambooth-library)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftuneavideo.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuzF6CTtjn-g), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuUlp1_ExsGQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fshowlab\u002FTune-A-Video\u002Fblob\u002Fmain\u002Fnotebooks\u002FTune-A-Video.ipynb) | 23.02.2023 |\n| GPEN | GAN Prior Embedded Network for Blind Face Restoration in the Wild | \u003Cul>\u003Cli>[Tao Yang](https:\u002F\u002Fcg.cs.tsinghua.edu.cn\u002Fpeople\u002F~tyang\u002F)\u003C\u002Fli> \u003Cli>[Peiran Ren](https:\u002F\u002Fscholar.google.com\u002Fcitations?&user=x5dEuxsAAAAJ)\u003C\u002Fli> \u003Cli>[Xuansong Xie](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=M0Ei1zkAAAAJ)\u003C\u002Fli> \u003Cli>[Lei Zhang](http:\u002F\u002Fwww4.comp.polyu.edu.hk\u002F~cslzhang\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyangxy\u002FGPEN?style=social)](https:\u002F\u002Fgithub.com\u002Fyangxy\u002FGPEN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.06070)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fvision.aliyun.com\u002Fexperience\u002Fdetail?spm=a211p3.14020179.J_7524944390.17.66cd4850wVDkUQ&tagName=facebody&children=EnhanceFace)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbiubug6\u002FPytorch_Retinaface), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyangxy\u002FGPEN\u002Fblob\u002Fmain\u002FGPEN.ipynb) | 15.02.2023 |\n| PyMAF-X | Кegression-based approach to recovering parametric full-body models from monocular images | \u003Cul>\u003Cli>[Hongwen Zhang](https:\u002F\u002Fhongwenzhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yating Tian](https:\u002F\u002Fgithub.com\u002Ftinatiansjz)\u003C\u002Fli> \u003Cli>[Yuxiang Zhang](https:\u002F\u002Fzhangyux15.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mengcheng Li](https:\u002F\u002Fgithub.com\u002FDw1010)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Liang An](https:\u002F\u002Fanl13.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhenan Sun](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fusers\u002Fznsun\u002F)\u003C\u002Fli> \u003Cli>[Yebin Liu](https:\u002F\u002Fwww.liuyebin.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_17890a89ad9c.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTPAMI.2023.3271691) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FHongwenZhang\u002FPyMAF-X?style=social)](https:\u002F\u002Fgithub.com\u002FHongwenZhang\u002FPyMAF-X) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.06400)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHongwenZhang\u002FDaNet-DensePose2SMPL), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FDensePose), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMicrosoft\u002Fhuman-pose-estimation.pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FMeshGraphormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fleoxiaobin\u002Fdeep-high-resolution-net.pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.liuyebin.com\u002Fpymaf-x\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FylOB0wCeV34)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F13Iytx1Hb0ZryEwbJdpXBW9ggDxs2Y-tL) | 14.02.2023 |\n| Disco Diffusion | A frankensteinian amalgamation of notebooks, models and techniques for the generation of AI Art and Animations | \u003Cul>\u003Cli>[Max Ingham](https:\u002F\u002Fgithub.com\u002Fsomnai-dreams)\u003C\u002Fli> \u003Cli>[Adam Letts](https:\u002F\u002Flinktr.ee\u002Fgandamu)\u003C\u002Fli> \u003Cli>[Daniel Russell](https:\u002F\u002Fgithub.com\u002Frusselldc)\u003C\u002Fli> \u003Cli>[Chigozie Nri](https:\u002F\u002Fgithub.com\u002Fchigozienri)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Falembics\u002Fdisco-diffusion?style=social)](https:\u002F\u002Fgithub.com\u002Falembics\u002Fdisco-diffusion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fguided-diffusion)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_DtWfh9oS54), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FgWxmtdZL8FE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyVJB6oD0_gM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Falembics\u002Fdisco-diffusion\u002Fblob\u002Fmain\u002FDisco_Diffusion.ipynb) | 11.02.2023 |\n| GrooVAE | Some applications of machine learning for generating and manipulating beats and drum performances | \u003Cul>\u003Cli>[Jon Gillick](https:\u002F\u002Fwww.jongillick.com\u002F)\u003C\u002Fli> \u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagenta\u002Fmagenta?style=social)](https:\u002F\u002Fgithub.com\u002Fmagenta\u002Fmagenta\u002Ftree\u002Fmain\u002Fmagenta\u002Fmodels\u002Fmusic_vae) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.06118)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fg.co\u002Fmagenta\u002Fgroovae)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fg.co\u002Fmagenta\u002Fgroove-datasets)\u003C\u002Fli>\u003Cli>[web app](https:\u002F\u002Fgroove-drums.glitch.me\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=x2YLmXzovDo)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fmagenta-demos\u002Fblob\u002Fmaster\u002Fcolab-notebooks\u002FGrooVAE.ipynb) | 02.02.2023 |\n| Multitrack MusicVAE | The models in this notebook are capable of encoding and decoding single measures of up to 8 tracks, optionally conditioned on an underlying chord | \u003Cul>\u003Cli>[Ian Simon](https:\u002F\u002Fgithub.com\u002Fiansimon)\u003C\u002Fli> \u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Colin Raffel](https:\u002F\u002Fcolinraffel.com\u002F\u002F)\u003C\u002Fli> \u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli> \u003Cli>[Douglas Eck](https:\u002F\u002Fgithub.com\u002Fdouglaseck)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1806.00195)\u003C\u002Fli>\u003Cli>[blog post](http:\u002F\u002Fg.co\u002Fmagenta\u002Fmultitrack)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmagenta\u002Fmagenta-demos\u002Fblob\u002Fmaster\u002Fcolab-notebooks\u002FMultitrack_MusicVAE.ipynb) | 02.02.2023 |\n| MusicVAE | A Hierarchical Latent Vector Model for Learning Long-Term Structure in Music | \u003Cul>\u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli> \u003Cli>[Colin Raffel](https:\u002F\u002Fcolinraffel.com\u002F\u002F)\u003C\u002Fli> \u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli> \u003Cli>[Douglas Eck](https:\u002F\u002Fgithub.com\u002Fdouglaseck)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1803.05428)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fg.co\u002Fmagenta\u002Fmusic-vae)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmagenta.tensorflow.org\u002Fmusic-vae)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLBUMAYA6kvGU8Cgqh709o5SUvo-zHGTxr)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmagenta\u002Fmagenta-demos\u002Fblob\u002Fmaster\u002Fcolab-notebooks\u002FMusicVAE.ipynb) | 02.02.2023 |\n| Learning to Paint | Learning to Paint With Model-based Deep Reinforcement Learning | [Manuel Romero](https:\u002F\u002Fmrm8488.github.io\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_6b548051740c.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV.2019.00880) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.04411)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Freinforcementlearning\u002Fcomments\u002Fb5lpfl\u002Flearning_to_paint_with_modelbased_deep\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=YmOgKZ5oipk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmrm8488\u002Fshared_colab_notebooks\u002Fblob\u002Fmaster\u002Fcustom_learningtopaint.ipynb) | 01.02.2023 |\n| LORA | Low-Rank Adaptation, which freezes the pre-trained model weights and injects trainable rank decomposition matrices into each layer of the Transformer architecture, greatly reducing the number of trainable parameters for downstream tasks | \u003Cul>\u003Cli>[Edward Hu](https:\u002F\u002Fedwardjhu.com\u002F)\u003C\u002Fli> \u003Cli>[Yelong Shen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=S6OFEFEAAAAJ)\u003C\u002Fli> \u003Cli>[Phillip Wallis](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=8IqHSXYAAAAJ)\u003C\u002Fli> \u003Cli>[Zeyuan Allen-Zhu](http:\u002F\u002Fzeyuan.allen-zhu.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yuanzhi Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=aHtfItQAAAAJ)\u003C\u002Fli> \u003Cli>[Shean Wang](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fshean-wang-18a20841)\u003C\u002Fli> \u003Cli>[Lu Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=JWJ_SNcAAAAJ)\u003C\u002Fli> \u003Cli>[Weizhu Chen](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fwzchen\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcloneofsimo\u002Flora?style=social)](https:\u002F\u002Fgithub.com\u002Fcloneofsimo\u002Flora) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.09685), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.12242), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.01618), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.05744), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.04647)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FLoRA)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Flora-library\u002FLoRA-DreamBooth-Training-UI), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Flora)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@shelikohan\u002Flow-rank-adapter-lora-explained-0d3677395639), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Ftowardsdatascience.com\u002Funderstanding-lora-low-rank-adaptation-for-finetuning-large-models-936bce1a07c6\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Floralib\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLocalLLaMA\u002Fcomments\u002F1diw5fb\u002Funderstanding_lora_a_5minute_visual_guide_to\u002F), [\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLocalLLaMA\u002Fcomments\u002F17z91wk\u002Fpractical_tips_for_finetuning_llms_using_lora\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdA-NhCtrrVE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Ft509sv5MT0w), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FKEv-F5UkhxU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDhRoTONcyZE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPXWYUTMt-AU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FX4VvO3G6_vw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1iSFDpRBKEWr2HLlz243rbym3J2X95kcy) | 30.01.2023 |\n| Instant-NGP | Instant Neural Graphics Primitives with a Multiresolution Hash Encoding | \u003Cul>\u003Cli>[Thomas Müller](https:\u002F\u002Ftom94.net\u002F)\u003C\u002Fli> \u003Cli>[Alex Evans](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Falex-evans)\u003C\u002Fli> \u003Cli>[Christoph Schied](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Fchristoph-schied)\u003C\u002Fli> \u003Cli>[Alexander Keller](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Falex-keller)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9bc55e3acb8c.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3528223.3530127) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNVlabs\u002Finstant-ngp?style=social)](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Finstant-ngp) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.05989)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fdeveloper.nvidia.com\u002Fblog\u002Fgetting-started-with-nvidia-instant-nerfs\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Ftiny-cuda-nn), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FIDLabMedia\u002Flarge-lightfields-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnickponline\u002Fdd-nerf-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Focornut\u002Fimgui), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnothings\u002Fstb)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnvlabs.github.io\u002Finstant-ngp\u002F)\u003C\u002Fli>\u003Cli>[tutorial](https:\u002F\u002Fwww.nvidia.com\u002Fen-us\u002Fon-demand\u002Fsession\u002Fsiggraph2022-sigg22-s-16\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fj8tMk-GE8hY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8GbENSmdVeE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDJ2hcC1orc4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fz3-fjYzd0BA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FNVlabs\u002Finstant-ngp\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finstant_ngp.ipynb) | 18.01.2023 |\n| Fourier Feature Networks | Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains | \u003Cul>\u003Cli>[Matthew Tancik](https:\u002F\u002Fwww.matthewtancik.com\u002F)\u003C\u002Fli> \u003Cli>[Pratul Srinivasan](https:\u002F\u002Fpratulsrinivasan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ben Mildenhall](https:\u002F\u002Fbmild.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sara Fridovich-Keil](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~sfk\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Nithin Raghavan](https:\u002F\u002Fcseweb.ucsd.edu\u002F\u002F~n2raghavan\u002F)\u003C\u002Fli> \u003Cli>[Utkarsh Singhal](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=lvA86MYAAAAJ)\u003C\u002Fli> \u003Cli>[Ravi Ramamoorthi](https:\u002F\u002Fcseweb.ucsd.edu\u002F\u002F~ravir\u002F)\u003C\u002Fli> \u003Cli>[Jon Barron](https:\u002F\u002Fjonbarron.info\u002F)\u003C\u002Fli> \u003Cli>[Ren Ng](https:\u002F\u002Fwww2.eecs.berkeley.edu\u002FFaculty\u002FHomepages\u002Fyirenng.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftancik\u002Ffourier-feature-networks?style=social)](https:\u002F\u002Fgithub.com\u002Ftancik\u002Ffourier-feature-networks) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1806.07572)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2020\u002Fhash\u002F55053683268957697aa39fba6f231c68-Abstract.html), [\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F2007\u002Fhash\u002F013a006f03dbc5392effeb8f18fda755-Abstract.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fbmild.github.io\u002Ffourfeat\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FnVA6K6Sn2S4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftancik\u002Ffourier-feature-networks\u002Fblob\u002Fmaster\u002FDemo.ipynb) | 17.01.2023 |\n| HybrIK | Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation | \u003Cul>\u003Cli>[Jiefeng Li](https:\u002F\u002Fjeffli.site\u002F)\u003C\u002Fli> \u003Cli>[Chao Xu](https:\u002F\u002Fwww.isdas.cn\u002F)\u003C\u002Fli> \u003Cli>[Zhicun Chen](https:\u002F\u002Fgithub.com\u002Fchenzhicun)\u003C\u002Fli> \u003Cli>[Siyuan Bian](https:\u002F\u002Fgithub.com\u002Fbiansy000)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lixin Yang](https:\u002F\u002Flixiny.github.io\u002F)\u003C\u002Fli> \u003Cli>[Cewu Lu](https:\u002F\u002Fwww.mvig.org\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_e4ff3e3e581a.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00339) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FJeff-sjtu\u002FHybrIK?style=social)](https:\u002F\u002Fgithub.com\u002FJeff-sjtu\u002FHybrIK) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.14672)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmks0601\u002F3DMPPE_POSENET_RELEASE)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fjeffli.site\u002FHybrIK\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002F3d-human-pose-estimation-on-3dpw?p=hybrik-a-hybrid-analytical-neural-inverse)\u003C\u002Fli>\u003Cli>[supp](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fsupplemental\u002FLi_HybrIK_A_Hybrid_CVPR_2021_supplemental.zip)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FtvwnXXH7xIw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1n41l7I2NxWseuruVQEU8he2XqzSXhu2f) | 01.01.2023 |\n| Composable-Diffusion | Method can generate scenes at test time that are substantially more complex than those seen in training, composing sentence descriptions, object relations, human facial attributes, and even generalizing to new combinations that are rarely seen in the real world | \u003Cul>\u003Cli>[Nan Liu](https:\u002F\u002Fnanliu.io)\u003C\u002Fli> \u003Cli>[Shuang Li](https:\u002F\u002Fshuangli01.github.io)\u003C\u002Fli> \u003Cli>[Yilun Du](https:\u002F\u002Fyilundu.github.io)\u003C\u002Fli> \u003Cli>[Antonio Torralba](https:\u002F\u002Fwww.csail.mit.edu\u002Fperson\u002Fantonio-torralba)\u003C\u002Fli> \u003Cli>[Joshua Tenenbaum](http:\u002F\u002Fweb.mit.edu\u002Fcocosci\u002Fjosh.html)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c36afe809fc9.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-19790-1_26) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fenergy-based-model\u002FCompositional-Visual-Generation-with-Composable-Diffusion-Models-PyTorch?style=social)](https:\u002F\u002Fgithub.com\u002Fenergy-based-model\u002FCompositional-Visual-Generation-with-Composable-Diffusion-Models-PyTorch) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.01714)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fpoint-e), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fimproved-diffusion)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FShuang59\u002FComposable-Diffusion)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fenergy-based-model.github.io\u002FCompositional-Visual-Generation-with-Composable-Diffusion-Models\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002Fxwplfv\u002Fand_prompt_combinations_just_landed_in\u002F), [\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002Fxf5jow\u002Fcompositional_diffusion\u002F), [\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002Fxoq7ik\u002Fcomposable_diffusion_a_new_development_to_greatly\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FmQzF6BDKes4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fenergy-based-model\u002FCompositional-Visual-Generation-with-Composable-Diffusion-Models-PyTorch\u002Fblob\u002Fmain\u002Fnotebooks\u002Fdemo.ipynb) | 23.12.2022 |\n| Score Jacobian Chaining | Apply chain rule on the learned gradients, and back-propagate the score of a diffusion model through the Jacobian of a differentiable renderer, which we instantiate to be a voxel radiance field | \u003Cul>\u003Cli>[Haochen Wang](https:\u002F\u002Fwhc.is\u002F)\u003C\u002Fli> \u003Cli>[Xiaodan Du](https:\u002F\u002Fxiaodan.io\u002F)\u003C\u002Fli> \u003Cli>[Jiahao Li](https:\u002F\u002Fjiahao.ai\u002F)\u003C\u002Fli> \u003Cli>[Raymond Yeh](https:\u002F\u002Fraymond-yeh.com\u002F)\u003C\u002Fli> \u003Cli>[Greg Shakhnarovich](https:\u002F\u002Fhome.ttic.edu\u002F~gregory\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a41f02ed2b74.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52729.2023.01214) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fpals-ttic\u002Fsjc?style=social)](https:\u002F\u002Fgithub.com\u002Fpals-ttic\u002Fsjc) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.00774), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.00364)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FMirageML\u002Fsjc)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpals.ttic.edu\u002Fp\u002Fscore-jacobian-chaining)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002Fzac8z4\u002Fscore_jacobian_chaining_lifting_pretrained_2d\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FMmDSLc6CjoI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F1oeruRLKoiU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1zixo66UYGl70VOPy053o7IV_YkQt5lCZ) | 05.12.2022 |\n| Demucs | Hybrid Spectrogram and Waveform Source Separation | [Alexandre Défossez](https:\u002F\u002Fai.honu.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fdemucs?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdemucs) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.03600), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.01733), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.05418), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1805.02410)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fadefossez\u002Fmdx21_demucs), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCarlGao4\u002FDemucs-Gui), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkuielab\u002Fmdx-net-submission), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ff90\u002FWave-U-Net)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1dC9nVxk3V_VPjUADsnFu8EiT-xnU1tGH) | 21.11.2022 |\n| FSGAN | Face Swapping GAN for face swapping and reenactment | \u003Cul>\u003Cli>[Yuval Nirkin](https:\u002F\u002Fnirkin.com\u002F)\u003C\u002Fli> \u003Cli>[Yosi Keller](https:\u002F\u002Fyosikeller.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tal Hassner](https:\u002F\u002Ftalhassner.github.io\u002Fhome\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_31e02e091914.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV.2019.00728) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYuvalNirkin\u002Ffsgan?style=social)](https:\u002F\u002Fgithub.com\u002FYuvalNirkin\u002Ffsgan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.05932)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fondyari\u002FFaceForensics)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fodsc.medium.com\u002Ffsgan-subject-agnostic-face-swapping-and-reenactment-2f033b0ea83c)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnirkin.com\u002Ffsgan\u002F), [project](https:\u002F\u002Fnirkin.com\u002Ffsgan-v2\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBsITEVX6hkE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fduo-tHbSdMk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FcfEqjXkCcCI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F4rNUkmEqgng)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FYuvalNirkin\u002Ffsgan\u002Fblob\u002Fmaster\u002Ffsgan\u002Finference\u002Fface_swapping.ipynb) | 16.11.2022 |\n| StyleCLIP | Text-Driven Manipulation of StyleGAN Imager | \u003Cul>\u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zongze Wu](https:\u002F\u002Fgithub.com\u002Fbetterze)\u003C\u002Fli> \u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli> \u003Cli>[Dani Lischinski](https:\u002F\u002Fpages.cs.huji.ac.il\u002Fdanix\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9cadabca3ad7.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00209) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Forpatashnik\u002FStyleCLIP?style=social)](https:\u002F\u002Fgithub.com\u002Forpatashnik\u002FStyleCLIP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.17249), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.12799)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5icI0NgALnQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPhR1gpXDu0w), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fd1OET63Ulwc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRAXrwPskNso)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Forpatashnik\u002FStyleCLIP\u002Fblob\u002Fmain\u002Fnotebooks\u002FStyleCLIP_global_torch.ipynb) | 30.10.2022 |\n| AST | Audio Spectrogram Transformer, the first convolution-free, purely attention-based model for audio classification | \u003Cul>\u003Cli>[Yuan Gong](https:\u002F\u002Fyuangongnd.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yu-An Chung](https:\u002F\u002Fiamyuanchung.github.io\u002F)\u003C\u002Fli> \u003Cli>[James Glass](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=pfGI-KcAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_bb66827c48fd.png)](https:\u002F\u002Fdoi.org\u002F10.21437\u002FInterspeech.2021-698) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYuanGongND\u002Fast?style=social)](https:\u002F\u002Fgithub.com\u002FYuanGongND\u002Fast) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.01778), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.06760), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2110.09784), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.01243)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYuanGongND\u002Fltu), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYuanGongND\u002Fssast), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYuanGongND\u002Fpsla)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Freview-ast-audio-spectrogram-transformer-a108a5775d2f)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FiKqmvNSGuyw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F9vGeIyeRRNQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FCSRDbqGY0Vw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FYuanGongND\u002Fast\u002Fblob\u002Fmaster\u002Fcolab\u002FAST_Inference_Demo.ipynb) | 18.10.2022 |\n| MotionDiffuse | The first diffusion model-based text-driven motion generation framework, which demonstrates several desired properties over existing methods | \u003Cul>\u003Cli>[Mingyuan Zhang](https:\u002F\u002Fmingyuan-zhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhongang Cai](https:\u002F\u002Fcaizhongang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liang Pan](https:\u002F\u002Fgithub.com\u002Fpaul007pl)\u003C\u002Fli> \u003Cli>[Fangzhou Hong](https:\u002F\u002Fhongfz16.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xinying Guo](https:\u002F\u002Fgxyes.github.io\u002F)\u003C\u002Fli> \u003Cli>[Lei Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jZH2IPYAAAAJ)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmingyuan-zhang\u002FMotionDiffuse?style=social)](https:\u002F\u002Fgithub.com\u002Fmingyuan-zhang\u002FMotionDiffuse) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.15001)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fmingyuan\u002FMotionDiffuse)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmingyuan-zhang.github.io\u002Fprojects\u002FMotionDiffuse.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FU5PTnw490SA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Dp6VsZp2ozKuu9ccMmsDjyij_vXfCYb3) | 13.10.2022 |\n| VToonify | Leverages the mid- and high-resolution layers of StyleGAN to render high-quality artistic portraits based on the multi-scale content features extracted by an encoder to better preserve the frame details | \u003Cul>\u003Cli>[Shuai Yang](https:\u002F\u002Fwilliamyang1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liming Jiang](https:\u002F\u002Fliming-jiang.com\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cba6c3f73678.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3550454.3555437) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwilliamyang1991\u002FVToonify?style=social)](https:\u002F\u002Fgithub.com\u002Fwilliamyang1991\u002FVToonify) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.11224), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2001.02890)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzllrunning\u002Fface-parsing.PyTorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhujiapeng\u002FLowRankGAN)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FPKUWilliamYang\u002FVToonify), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FPKUWilliamYang\u002FVToonify\u002Ftree\u002Fmain\u002Fmodels)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fproject\u002Fvtoonify\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0_OmVhDgYuY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](http:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fwilliamyang1991\u002FVToonify\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 07.10.2022 |\n| PyMAF | Pyramidal Mesh Alignment Feedback loop in regression network for well-aligned body mesh recovery and extend it for the recovery of expressive full-body models | \u003Cul>\u003Cli>[Hongwen Zhang](https:\u002F\u002Fhongwenzhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yating Tian](https:\u002F\u002Fgithub.com\u002Ftinatiansjz)\u003C\u002Fli> \u003Cli>[Yuxiang Zhang](https:\u002F\u002Fzhangyux15.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mengcheng Li](https:\u002F\u002Fgithub.com\u002FDw1010)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Liang An](https:\u002F\u002Fanl13.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhenan Sun](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fusers\u002Fznsun\u002F)\u003C\u002Fli> \u003Cli>[Yebin Liu](https:\u002F\u002Fwww.liuyebin.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FHongwenZhang\u002FPyMAF?style=social)](https:\u002F\u002Fgithub.com\u002FHongwenZhang\u002FPyMAF) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.06400), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.16507)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Feft), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHongwenZhang\u002FDaNet-DensePose2SMPL), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FDensePose), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMicrosoft\u002Fhuman-pose-estimation.pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.liuyebin.com\u002Fpymaf-x\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyqEmznSKjYI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FylOB0wCeV34)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F11RXLsH9BdoSCwY6G-IX7KgqDxVoImu6K) | 06.10.2022 |\n| AlphaTensor | Discovering faster matrix multiplication algorithms with reinforcement learning | \u003Cul>\u003Cli>[Alhussein Fawzi](http:\u002F\u002Fwww.alhusseinfawzi.info\u002F)\u003C\u002Fli> \u003Cli>[Matej Balog](http:\u002F\u002Fmatejbalog.eu\u002F)\u003C\u002Fli> \u003Cli>[Aja Huang](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FAja_Huang)\u003C\u002Fli> \u003Cli>[Thomas Hubert](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=WXG0QfMAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Bernardino Romera-Paredes](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fromeraparedes\u002F)\u003C\u002Fli> \u003Cli>[Mohammadamin Barekatain](http:\u002F\u002Fbarekatain.me\u002F)\u003C\u002Fli> \u003Cli>[Alexander Novikov](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jMUkLqwAAAAJ)\u003C\u002Fli> \u003Cli>[Francisco Ruiz](https:\u002F\u002Ffranrruiz.github.io\u002F)\u003C\u002Fli> \u003Cli>[Julian Schrittwieser](https:\u002F\u002Fwww.furidamu.org\u002F)\u003C\u002Fli> \u003Cli>[Grzegorz Swirszcz](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fgrzegorzswirszcz\u002Fhome)\u003C\u002Fli> \u003Cli>[David Silver](https:\u002F\u002Fwww.davidsilver.uk\u002F)\u003C\u002Fli> \u003Cli>[Demis Hassabis](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FDemis_Hassabis)\u003C\u002Fli> \u003Cli>[Pushmeet Kohli](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fpushmeet\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_37c21b7bb6c8.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41586-022-05172-4) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Falphatensor?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Falphatensor) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fwww.deepmind.com\u002Fblog\u002Fdiscovering-novel-algorithms-with-alphatensor)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3N3Bl5AA5QU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FgpYnDls4PdQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIYgZS2EvnLI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8ILk4Wjo5rc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Falphatensor\u002Fblob\u002Fmaster\u002Fnonequivalence\u002Finspect_factorizations_notebook.ipynb) | 04.10.2022 |\n| Swin2SR | Novel Swin Transformer V2, to improve SwinIR for image super-resolution, and in particular, the compressed input scenario | \u003Cul>\u003Cli>[Marcos Conde](https:\u002F\u002Fmv-lab.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ui-Jin Choi](https:\u002F\u002Fgithub.com\u002FChoiuijin1125)\u003C\u002Fli> \u003Cli>[Maxime Burchi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=7S_l2eAAAAAJ)\u003C\u002Fli> \u003Cli>[Radu Timofte](https:\u002F\u002Fwww.informatik.uni-wuerzburg.de\u002Fcomputervision\u002Fhome\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_18e61c523853.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-25063-7_42) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmv-lab\u002Fswin2sr?style=social)](https:\u002F\u002Fgithub.com\u002Fmv-lab\u002Fswin2sr) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.11345), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.10257), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.11184), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.09883)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcszn\u002FKAIR\u002F), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmv-lab\u002FAISP), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FSwin-Transformer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fjjourney1125\u002Fswin2sr)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fcode\u002Fjesucristo\u002Fsuper-resolution-demo-swin2sr-official\u002F), [\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fdatasets\u002Fjesucristo\u002Fsuper-resolution-benchmarks), [\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fjinssaa\u002Fofficial-swin2sr-demo-results\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1paPrt62ydwLv2U2eZqfcFsePI4X4WRR1) | 03.10.2022 |\n| Functa | From data to functa: Your data point is a function and you can treat it like one | \u003Cul>\u003Cli>[Emilien Dupont](https:\u002F\u002Femiliendupont.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hyunjik Kim](https:\u002F\u002Fhyunjik11.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ali Eslami](http:\u002F\u002Farkitus.com\u002F)\u003C\u002Fli> \u003Cli>[Danilo Rezende](https:\u002F\u002Fdanilorezende.com\u002Fabout\u002F)\u003C\u002Fli> \u003Cli>[Dan Rosenbaum](https:\u002F\u002Fdanrsm.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepmind\u002Ffuncta?style=social)](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Ffuncta) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.12204)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsxyu\u002Fpixel-nerf), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fjaxline)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fdatasets\u002Fcatalog\u002Fceleb_a_hq)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Ffuncta\u002Fblob\u002Fmain\u002Fmodulation_visualization_colab.ipynb) | 24.09.2022 |\n| DeOldify (photo) | Colorize your own photos! | \u003Cul>\u003Cli>[Jason Antic](https:\u002F\u002Fgithub.com\u002Fjantic)\u003C\u002Fli> \u003Cli>[Matt Robinson](https:\u002F\u002Fgithub.com\u002Fmc-robinson)\u003C\u002Fli> \u003Cli>[María Benavente](https:\u002F\u002Fgithub.com\u002Fmariabg)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjantic\u002FDeOldify?style=social)](https:\u002F\u002Fgithub.com\u002Fjantic\u002FDeOldify) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1805.08318), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.08500)\u003C\u002Fli>\u003Cli>[model](https:\u002F\u002Fdata.deepai.org\u002Fdeoldify\u002FColorizeArtistic_gen.pth)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FTheWayWeWere\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002FDeOldify)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fdeoldify.ai\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fjantic\u002FDeOldify\u002Fblob\u002Fmaster\u002FImageColorizerColab.ipynb) | 19.09.2022 |\n| DeOldify (video) | Colorize your own videos! | [Jason Antic](https:\u002F\u002Fgithub.com\u002Fjantic) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjantic\u002FDeOldify?style=social)](https:\u002F\u002Fgithub.com\u002Fjantic\u002FDeOldify) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1805.08318), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.08500)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Felement-ai-research-lab\u002Fstabilizing-neural-style-transfer-for-video-62675e203e42)\u003C\u002Fli>\u003Cli>[model](https:\u002F\u002Fdata.deepai.org\u002Fdeoldify\u002FColorizeVideo_gen.pth)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FNickelodeons\u002F), [\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fsilentmoviegifs\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002FDeOldify)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fdeoldify.ai\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](http:\u002F\u002Fwww.youtube.com\u002Fwatch?v=l3UXXid04Ys), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](http:\u002F\u002Fwww.youtube.com\u002Fwatch?v=EXn-n2iqEjI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fjantic\u002FDeOldify\u002Fblob\u002Fmaster\u002FVideoColorizerColab.ipynb) | 19.09.2022 |\n| Real-ESRGAN | Extend the powerful ESRGAN to a practical restoration application, which is trained with pure synthetic data | \u003Cul>\u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liangbin Xie](https:\u002F\u002Fliangbinxie.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chao Dong](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=OSDCB0UAAAAJ)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c3c44214a50a.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCVW54120.2021.00217) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fxinntao\u002FReal-ESRGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fxinntao\u002FReal-ESRGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.10833)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002FESRGAN), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002Ffacexlib), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002FHandyView), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTencent\u002Fncnn), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnihui\u002Fwaifu2x-ncnn-vulkan)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1k2Zod6kSHEvraybHl50Lys0LerhyTMCo) | 18.09.2022 |\n| IDE-3D | Interactive Disentangled Editing for High-Resolution 3D-aware Portrait Synthesis | \u003Cul>\u003Cli>[Jingxiang Sun](https:\u002F\u002Fmrtornado24.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xuan Wang](https:\u002F\u002Fxuanwangvc.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yichun Shi](https:\u002F\u002Fseasonsh.github.io\u002F)\u003C\u002Fli> \u003Cli>[Lizhen Wang](https:\u002F\u002Flizhenwangt.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jue Wang](https:\u002F\u002Fjuewang725.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yebin Liu](http:\u002F\u002Fwww.liuyebin.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_952a6250d74e.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3550454.3555506) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FMrTornado24\u002FIDE-3D?style=social)](https:\u002F\u002Fgithub.com\u002FMrTornado24\u002FIDE-3D) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.15517), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Feg3d), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan3)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FKj5XY_J2Alk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FMrTornado24\u002FIDE-3D\u002Fblob\u002Fmain\u002Finversion\u002Fnotebooks\u002Finference_playground.ipynb) | 08.09.2022 |\n| Decision Transformers | An architecture that casts the problem of RL as conditional sequence modeling | \u003Cul>\u003Cli>[Lili Chen](http:\u002F\u002Fwww.lilichen.me\u002F)\u003C\u002Fli> \u003Cli>[Kevin Lu](https:\u002F\u002Fkzl.github.io\u002F)\u003C\u002Fli> \u003Cli>[Aravind Rajeswaran](https:\u002F\u002Faravindr93.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kimin Lee](https:\u002F\u002Fsites.google.com\u002Fview\u002Fkiminlee)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Aditya Grover](https:\u002F\u002Faditya-grover.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michael Laskin](https:\u002F\u002Fwww.mishalaskin.com\u002F)\u003C\u002Fli> \u003Cli>[Pieter Abbeel](http:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~pabbeel\u002F)\u003C\u002Fli> \u003Cli>[Aravind Srinivas](https:\u002F\u002Fgithub.com\u002Faravindsrinivas)\u003C\u002Fli> \u003Cli>[Igor Mordatch](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Vzr1RukAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fkzl\u002Fdecision-transformer?style=social)](https:\u002F\u002Fgithub.com\u002Fkzl\u002Fdecision-transformer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.01345)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fmodels?other=gym-continous-control), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fedbeeching\u002Fdecision-transformer-gym-hopper-expert), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Fdecision_transformer)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsites.google.com\u002Fberkeley.edu\u002Fdecision-transformer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FAutoregressive_model)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fk08N5a0gG0A), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-buULmf7dec), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F83QN9S-0I84), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fw4Bw8WYL8Ps)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1K3UuajwoPY1MzRKNkONNRS3gS5DxZ-qF) | 06.09.2022 |\n| textual-inversion | An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion | \u003Cul>\u003Cli>[Rinon Gal](https:\u002F\u002Frinongal.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuval Atzmon](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Fyuval-atzmon)\u003C\u002Fli> \u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Amit Bermano](https:\u002F\u002Fwww.cs.tau.ac.il\u002F~amberman\u002F)\u003C\u002Fli> \u003Cli>[Gal Chechik](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Fgal-chechik)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Frinongal\u002Ftextual_inversion?style=social)](https:\u002F\u002Fgithub.com\u002Frinongal\u002Ftextual_inversion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.01618)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftextual-inversion.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Ff3oXa7_SYek), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FopD_H9bED9Y)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Frinongal\u002Ftextual_inversion\u002Fblob\u002Fmaster\u002Fscripts\u002Flatent_imagenet_diffusion.ipynb) | 21.08.2022 |\n| StyleGAN-Human | A Data-Centric Odyssey of Human Generation | \u003Cul>\u003Cli>[Jianglin Fu](https:\u002F\u002Fgithub.com\u002FarleneF)\u003C\u002Fli> \u003Cli>[Shikai Li](https:\u002F\u002Fgithub.com\u002Fleeskyed)\u003C\u002Fli> \u003Cli>[Yuming Jiang](https:\u002F\u002Fyumingj.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kwan-Yee Lin](https:\u002F\u002Fkwanyeelin.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chen Qian](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=AerkT0YAAAAJ)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli> \u003Cli>[Wayne Wu](https:\u002F\u002Fwywu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_19bf517b7e29.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-19787-1_1) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fstylegan-human\u002Fstylegan-human?style=social)](https:\u002F\u002Fgithub.com\u002Fstylegan-human\u002Fstylegan-human) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.11823)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan3)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fstylegan-human.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fdataset\u002Fmarket-1501)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FnIrb9hwsdcI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F86b49sCz0Gg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fg3nmM6MdxwY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fp2uwqh_SFL8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1sgxoDM55iM07FS54vz9ALg1XckiYA2On) | 19.08.2022 |\n| Make-A-Scene | Scene-Based Text-to-Image Generation with Human Priors | \u003Cul>\u003Cli>[Oran Gafni](https:\u002F\u002Fgithub.com\u002Fogafni)\u003C\u002Fli> \u003Cli>[Adam Polyak](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=CP62OTMAAAAJ)\u003C\u002Fli> \u003Cli>[Oron Ashual](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=CUA9JCkAAAAJ)\u003C\u002Fli> \u003Cli>[Shelly Sheynin](https:\u002F\u002Fgithub.com\u002Fshellysheynin)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Devi Parikh](https:\u002F\u002Ffaculty.cc.gatech.edu\u002F~parikh\u002F)\u003C\u002Fli> \u003Cli>[Yaniv Taigman](https:\u002F\u002Fytaigman.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCasualGANPapers\u002FMake-A-Scene?style=social)](https:\u002F\u002Fgithub.com\u002FCasualGANPapers\u002FMake-A-Scene) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.13131)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FZM06MjPdoxw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1SPyQ-epTsAOAu8BEohUokN4-b5RM_TnE) | 12.08.2022 |\n| StyleGAN-NADA | Zero-Shot non-adversarial domain adaptation of pre-trained generators | \u003Cul>\u003Cli>[Rinon Gal](https:\u002F\u002Frinongal.github.io\u002F)\u003C\u002Fli> \u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haggai Maron](https:\u002F\u002Fhaggaim.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gal Chechik](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Fgal-chechik)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0cd2480f3d79.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3528223.3530164) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Frinongal\u002FStyleGAN-nada?style=social)](https:\u002F\u002Fgithub.com\u002Frinongal\u002FStyleGAN-nada) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.00946), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.17249), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.02699)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch\u002F), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fstylegan-nada.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Frinongal\u002Fstylegan-nada\u002Fblob\u002Fmain\u002Fstylegan_nada.ipynb) | 09.08.2022 |\n| YOLOv7 | Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors | \u003Cul>\u003Cli>[Chien-Yao Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=DkQh4M4AAAAJ)\u003C\u002Fli> \u003Cli>[Alexey Bochkovskiy](http:\u002F\u002Fwww.alexeyab.com\u002F)\u003C\u002Fli> \u003Cli>[Mark Liao](https:\u002F\u002Fwww.iis.sinica.edu.tw\u002Fpages\u002Fliao\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FWongKinYiu\u002Fyolov7?style=social)](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002Fyolov7) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.02696)\u003C\u002Fli>\u003Cli>[data](http:\u002F\u002Fimages.cocodataset.org\u002Fannotations\u002Fannotations_trainval2017.zip), [data](http:\u002F\u002Fimages.cocodataset.org\u002Fzips\u002Ftrain2017.zip), [data](http:\u002F\u002Fimages.cocodataset.org\u002Fzips\u002Fval2017.zip), [data](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002Fyolov7\u002Freleases\u002Fdownload\u002Fv0.1\u002Fcoco2017labels-segments.zip)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002Fyolor), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002FPyTorch_YOLOv4), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002FScaledYOLOv4), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMegvii-BaseDetection\u002FYOLOX), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDingXiaoH\u002FRepVGG), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJUGGHM\u002FOREPA_CVPR2022), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTexasInstruments\u002Fedgeai-yolov5\u002Ftree\u002Fyolo-pose)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Freal-time-object-detection-on-coco?p=yolov7-trainable-bag-of-freebies-sets-new)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL_Nji0JOuXg2QMohGK7wfzgJ-MavzXRHW), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-QWxJ0j9EY8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FWongKinYiu\u002Fyolov7\u002Fblob\u002Fmain\u002Ftools\u002Fcompare_YOLOv7_vs_YOLOv5m6_half.ipynb) | 09.08.2022 |\n| GLIP | Grounded language-image pre-training model for learning object-level, language-aware, and semantic-rich visual representations | \u003Cul>\u003Cli>[Liunian Harold Li](https:\u002F\u002Fliunian-harold-li.github.io\u002F)\u003C\u002Fli> \u003Cli>[Pengchuan Zhang](https:\u002F\u002Fpzzhang.github.io\u002Fpzzhang\u002F)\u003C\u002Fli> \u003Cli>[Haotian Zhang](https:\u002F\u002Fhaotian-zhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jianwei Yang](https:\u002F\u002Fjwyang.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chunyuan Li](https:\u002F\u002Fchunyuan.li\u002F)\u003C\u002Fli> \u003Cli>[Yiwu Zhong](https:\u002F\u002Fpages.cs.wisc.edu\u002F~yiwuzhong\u002F)\u003C\u002Fli> \u003Cli>[Lijuan Wang](https:\u002F\u002Fgithub.com\u002FLijuanWang)\u003C\u002Fli> \u003Cli>[Lu Yuan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=k9TsUVsAAAAJ)\u003C\u002Fli> \u003Cli>[Lei Zhang](https:\u002F\u002Fwww.leizhang.org\u002F)\u003C\u002Fli> \u003Cli>[Jenq-Neng Hwang](https:\u002F\u002Fpeople.ece.uw.edu\u002Fhwang\u002F)\u003C\u002Fli> \u003Cli>[Kai-Wei Chang](http:\u002F\u002Fweb.cs.ucla.edu\u002F~kwchang\u002F)\u003C\u002Fli> \u003Cli>[Jianfeng Gao](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fjfgao\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_085ff51929b8.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01069) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002FGLIP?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FGLIP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.03857), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.05836), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.01066), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.08790)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fproject\u002Fproject-florence-vl\u002Farticles\u002Fobject-detection-in-the-wild-via-grounded-language-image-pre-training\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgligen\u002FGLIGEN)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fharold\u002FGLIP)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Fglip-grounded-language-image-pre-training-2be2483295b3), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Ftowardsdatascience.com\u002Fglip-introducing-language-image-pre-training-to-object-detection-5ddb601873aa)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fzu1BGQBI4dU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F12x7v-_miN7-SRiziK3Cx4ffJzstBJNqb) | 30.07.2022 |\n| Anycost GAN | Interactive natural image editing | \u003Cul>\u003Cli>[Ji Lin](http:\u002F\u002Flinji.me\u002F)\u003C\u002Fli> \u003Cli>[Richard Zhang](https:\u002F\u002Frichzhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Frieder Ganz](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=u9ySZkUAAAAJ)\u003C\u002Fli> \u003Cli>[Song Han](https:\u002F\u002Fsonghan.mit.edu\u002F)\u003C\u002Fli> \u003Cli>[Jun-Yan Zhu](https:\u002F\u002Fwww.cs.cmu.edu\u002F~junyanz\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a5493ea31851.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01474) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmit-han-lab\u002Fanycost-gan?style=social)](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fanycost-gan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.03243)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fswitchablenorms\u002FCelebAMask-HQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffyu\u002Flsun)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fhanlab.mit.edu\u002Fprojects\u002Fanycost-gan\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=_yEziPl9AkM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmit-han-lab\u002Fanycost-gan\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fintro_colab.ipynb) | 20.07.2022 |\n| GFPGAN | Towards Real-World Blind Face Restoration with Generative Facial Prior | \u003Cul>\u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yu Li](https:\u002F\u002Fyu-li.github.io\u002F)\u003C\u002Fli> \u003Cli>[Honglun Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=KjQLROoAAAAJ)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_b133ac9e676e.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00905) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTencentARC\u002FGFPGAN?style=social)](https:\u002F\u002Fgithub.com\u002FTencentARC\u002FGFPGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2101.04061)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002Ffacexlib), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002FHandyView), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fxinntao.github.io\u002Fprojects\u002Fgfpgan)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1sVsoBd9AjckIXThgtZhGrHRfFI6UUYOo) | 13.07.2022 |\n| EPro-PnP | Generalized End-to-End Probabilistic Perspective-n-Points for Monocular Object Pose Estimation | \u003Cul>\u003Cli>[Hansheng Chen](https:\u002F\u002Flakonik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Pichao Wang](https:\u002F\u002Fwangpichao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Fan Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=WCRGTHsAAAAJ)\u003C\u002Fli> \u003Cli>[Wei Tian](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=aYKQn88AAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lu Xiong](https:\u002F\u002Fieeexplore.ieee.org\u002Fauthor\u002F37401835800)\u003C\u002Fli> \u003Cli>[Hao Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=pHN-QIwAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fafa450d6826.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTPAMI.2024.3354997) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftjiiv-cprg\u002FEPro-PnP?style=social)](https:\u002F\u002Fgithub.com\u002Ftjiiv-cprg\u002FEPro-PnP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.13254)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmegvii-research\u002Fpetr), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHuangJunJie2017\u002FBEVDet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffudan-zvg\u002FPolarFormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhiqi-li\u002FBEVFormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopen-mmlab\u002Fmmdetection3d)\u003C\u002Fli>\u003Cli>[nuScenes](https:\u002F\u002Fwww.nuscenes.org\u002Fobject-detection?externalData=no&mapData=no&modalities=Camera)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTonBodQ6EUU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftjiiv-cprg\u002FEPro-PnP\u002Fblob\u002Fmain\u002Fdemo\u002Ffit_identity.ipynb) | 12.07.2022 |\n| Text2Human | Text-driven controllable framework for a high-quality and diverse human generation | \u003Cul>\u003Cli>[Yuming Jiang](https:\u002F\u002Fyumingj.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shuai Yang](https:\u002F\u002Fwilliamyang1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haonan Qiu](http:\u002F\u002Fhaonanqiu.com\u002F)\u003C\u002Fli> \u003Cli>[Wayne Wu](https:\u002F\u002Fwywu.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_898be9756148.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3528223.3530104) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyumingj\u002FText2Human?style=social)](https:\u002F\u002Fgithub.com\u002Fyumingj\u002FText2Human) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.15996)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyumingj\u002FDeepFashion-MultiModal), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsamb-t\u002Funleashing-transformers)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fhysts\u002FText2Human), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FCVPR\u002Fdrawings-to-human)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyumingj.github.io\u002Fprojects\u002FText2Human.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyKh4VORA_E0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRV-g5BlH3Zg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1AVwbqLwMp_Gz3KTCgBTtnGVtXIlCZDPk) | 04.07.2022 |\n| VQ-Diffusion | Based on a VQ-VAE whose latent space is modeled by a conditional variant of the recently developed Denoising Diffusion Probabilistic Model | \u003Cul>\u003Cli>[Shuyang Gu](https:\u002F\u002Fgithub.com\u002Fcientgu)\u003C\u002Fli> \u003Cli>[Dong Chen](http:\u002F\u002Fwww.dongchen.pro\u002F)\u003C\u002Fli> \u003Cli>[Jianmin Bao](https:\u002F\u002Fjianminbao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Fang Wen](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Ffangwen\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Bo Zhang](https:\u002F\u002Fbo-zhang.me\u002F)\u003C\u002Fli> \u003Cli>[Dongdong Chen](http:\u002F\u002Fwww.dongdongchen.bid\u002F)\u003C\u002Fli> \u003Cli>[Lu Yuan](https:\u002F\u002Fscholar.google.com\u002Fcitations?&user=k9TsUVsAAAAJ)\u003C\u002Fli> \u003Cli>[Baining Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=h4kYmRYAAAAJ)\u003C\u002Fli> \u003Cli>[Shuyang Gu](https:\u002F\u002Fgithub.com\u002Fcientgu)\u003C\u002Fli> \u003Cli>[Zhicong Tang](https:\u002F\u002Fgithub.com\u002Fzzctan)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_3b7ddc97ebfd.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01043) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002FVQ-Diffusion?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FVQ-Diffusion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.14822), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.16007)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fehoogeboom\u002Fmultinomial_diffusion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fimproved-diffusion)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Ws0_wK2cnsWEnfB7HtmPT4bjCPElb40C) | 30.06.2022 |\n| OPT | Open Pre-trained Transformers is a family of NLP models trained on billions of tokens of text obtained from the internet | \u003Cul>\u003Cli>[Susan Zhang](https:\u002F\u002Fgithub.com\u002Fsuchenzang)\u003C\u002Fli> \u003Cli>[Stephen Roller](https:\u002F\u002Fstephenroller.com\u002F)\u003C\u002Fli> \u003Cli>[Naman Goyal](https:\u002F\u002Fgithub.com\u002Fngoyal2707)\u003C\u002Fli> \u003Cli>[Mikel Artetxe](https:\u002F\u002Fgithub.com\u002Fartetxem)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Moya Chen](https:\u002F\u002Fmoyachen.com\u002F)\u003C\u002Fli> \u003Cli>[Christopher Dewan](https:\u002F\u002Fgithub.com\u002Fm3rlin45)\u003C\u002Fli> \u003Cli>[Mona Diab](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=-y6SIhQAAAAJ)\u003C\u002Fli> \u003Cli>[Xi Victoria Lin](http:\u002F\u002Fvictorialin.net\u002F)\u003C\u002Fli> \u003Cli>[Todor Mihaylov](https:\u002F\u002Fgithub.com\u002Ftbmihailov)\u003C\u002Fli> \u003Cli>[Myle Ott](https:\u002F\u002Fmyleott.com\u002F)\u003C\u002Fli> \u003Cli>[Sam Shleifer](https:\u002F\u002Fgithub.com\u002Fsshleifer)\u003C\u002Fli> \u003Cli>[Kurt Shuster](https:\u002F\u002Fgithub.com\u002Fklshuster)\u003C\u002Fli> \u003Cli>[Daniel Simig](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=TtWU9fsAAAAJ)\u003C\u002Fli> \u003Cli>[Punit Singh Koura](https:\u002F\u002Fgithub.com\u002Fpunitkoura)\u003C\u002Fli> \u003Cli>[Anjali Sridhar](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fanjalisridhar\u002F)\u003C\u002Fli> \u003Cli>[Tianlu Wang](https:\u002F\u002Ftianlu-wang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Luke Zettlemoyer](https:\u002F\u002Fwww.cs.washington.edu\u002Fpeople\u002Ffaculty\u002Flsz\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fmetaseq?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fmetaseq\u002Ftree\u002Fmain\u002Fprojects\u002FOPT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.01068), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.02243), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.10350), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.11990)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fai.facebook.com\u002Fblog\u002Fdemocratizing-access-to-large-scale-language-models-with-opt-175b\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FEjg0OunCi9U)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F14wnxMvD9zsiBQo2FtTpxn6w2cpXCcb-7) | 29.06.2022 |\n| Customizing a Transformer Encoder | We will learn how to customize the encoder to employ new network architectures | [Chen Chen](https:\u002F\u002Fgithub.com\u002FchenGitHuber) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Fmodels?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fmodels\u002Ftree\u002Fmaster\u002Fofficial\u002Fnlp\u002Fmodeling) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.03762)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fmodels\u002Fblob\u002Fmaster\u002Fofficial\u002Fnlp\u002Fmodeling\u002Fnetworks\u002Fencoder_scaffold.py)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fmodels\u002Fblob\u002Fmaster\u002Fofficial\u002Fcolab\u002Fnlp\u002Fcustomize_encoder.ipynb) | 22.06.2022 |\n| MTTR | End-to-End Referring Video Object Segmentation with Multimodal Transformers | \u003Cul>\u003Cli>[Adam Botach](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fadam-botach)\u003C\u002Fli> \u003Cli>[Evgenii Zheltonozhskii](https:\u002F\u002Fevgeniizh.com\u002F)\u003C\u002Fli> \u003Cli>[Chaim Baskin](https:\u002F\u002Fgithub.com\u002Fchaimbaskin)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ba7fa64376b6.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00493) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmttr2021\u002FMTTR?style=social)](https:\u002F\u002Fgithub.com\u002Fmttr2021\u002FMTTR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.14821), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.11692), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.13230)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FSwinTransformer\u002FVideo-Swin-Transformer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FMTTR\u002FMTTR-Referring-Video-Object-Segmentation)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYqlhXgq6hcs)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F12p0jpSx3pJNfZk-y_L44yeHZlhsKVra-) | 20.06.2022 |\n| SwinIR | Image Restoration Using Swin Transformer | \u003Cul>\u003Cli>[Jingyun Liang](https:\u002F\u002Fjingyunliang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiezhang Cao](https:\u002F\u002Fgithub.com\u002Fcaojiezhang)\u003C\u002Fli> \u003Cli>[Guolei Sun](https:\u002F\u002Fgithub.com\u002FGuoleiSun)\u003C\u002Fli> \u003Cli>[Kai Zhang](https:\u002F\u002Fcszn.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Luc Van Gool](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=TwMib_QAAAAJ)\u003C\u002Fli> \u003Cli>[Radu Timofte](https:\u002F\u002Fwww.informatik.uni-wuerzburg.de\u002Fcomputervision\u002Fhome\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_59e73815c1c5.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCVW54120.2021.00210) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FJingyunLiang\u002FSwinIR?style=social)](https:\u002F\u002Fgithub.com\u002FJingyunLiang\u002FSwinIR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.10257), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.10833)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcszn\u002FBSRGAN), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FSwin-Transformer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcszn\u002FKAIR)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgist\u002FJingyunLiang\u002Fa5e3e54bc9ef8d7bf594f6fee8208533\u002Fswinir-demo-on-real-world-image-sr.ipynb) | 17.06.2022 |\n| VRT | A Video Restoration Transformer | \u003Cul>\u003Cli>[Jingyun Liang](https:\u002F\u002Fjingyunliang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiezhang Cao](https:\u002F\u002Fgithub.com\u002Fcaojiezhang)\u003C\u002Fli> \u003Cli>[Yuchen Fan](https:\u002F\u002Fychfan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kai Zhang](https:\u002F\u002Fcszn.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yawei Li](https:\u002F\u002Fofsoundof.github.io\u002F)\u003C\u002Fli> \u003Cli>[Radu Timofte](https:\u002F\u002Fwww.informatik.uni-wuerzburg.de\u002Fcomputervision\u002Fhome\u002F)\u003C\u002Fli> \u003Cli>[Luc Van Gool](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=TwMib_QAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_aedd82489829.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTIP.2024.3372454) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FJingyunLiang\u002FVRT?style=social)](https:\u002F\u002Fgithub.com\u002FJingyunLiang\u002FVRT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.12288)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcszn\u002FKAIR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FSwinTransformer\u002FVideo-Swin-Transformer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopen-mmlab\u002Fmmediting)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgist\u002FJingyunLiang\u002Fdeb335792768ad9eb73854a8efca4fe0\u002Fvrt-demo-on-video-restoration.ipynb) | 15.06.2022 |\n| Omnivore | A single model which excels at classifying images, videos, and single-view 3D data using exactly the same model parameters | \u003Cul>\u003Cli>[Rohit Girdhar](http:\u002F\u002Frohitgirdhar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mannat Singh](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QOO8OCcAAAAJ)\u003C\u002Fli> \u003Cli>[Nikhila Ravi](https:\u002F\u002Fnikhilaravi.com\u002F)\u003C\u002Fli> \u003Cli>[Laurens Maaten](https:\u002F\u002Flvdmaaten.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Armand Joulin](https:\u002F\u002Fai.facebook.com\u002Fpeople\u002Farmand-joulin\u002F)\u003C\u002Fli> \u003Cli>[Ishan Misra](https:\u002F\u002Fimisra.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_51a58de9ffd7.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01563) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fomnivore?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fomnivore) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.08377), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.08356)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002Fomnivore)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ffacebookresearch.github.io\u002Fomnivore\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fdataset\u002Fepic-kitchens-100)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fomnivore\u002Fblob\u002Fmain\u002Finference_tutorial.ipynb) | 14.06.2022 |\n| Dream Fields | Zero-Shot Text-Guided Object Generation | \u003Cul>\u003Cli>[Ajay Jain](https:\u002F\u002Fajayj.com\u002F)\u003C\u002Fli> \u003Cli>[Ben Mildenhall](https:\u002F\u002Fbmild.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jon Barron](https:\u002F\u002Fjonbarron.info\u002F)\u003C\u002Fli> \u003Cli>[Pieter Abbeel](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~pabbeel\u002F)\u003C\u002Fli> \u003Cli>[Ben Poole](https:\u002F\u002Fcs.stanford.edu\u002F~poole\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_d53812bab803.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00094) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fgoogle-research?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fgoogle-research\u002Ftree\u002Fmaster\u002Fdreamfields) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.01455), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.00677), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.13415)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fajayjain\u002FDietNeRF), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmipnerf)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fajayj.com\u002Fdreamfields)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F1Fke6w46tv4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F17GtPqdUCbG5CsmTnQFecPpoq_zpNKX7A) | 10.06.2022 |\n| Detic | Detecting Twenty-thousand Classes using Image-level Supervision | \u003Cul>\u003Cli>[Xingyi Zhou](https:\u002F\u002Fwww.cs.utexas.edu\u002F~zhouxy\u002F)\u003C\u002Fli> \u003Cli>[Rohit Girdhar](https:\u002F\u002Frohitgirdhar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Armand Joulin](https:\u002F\u002Fai.facebook.com\u002Fpeople\u002Farmand-joulin\u002F)\u003C\u002Fli> \u003Cli>[Philipp Krähenbühl](https:\u002F\u002Fgithub.com\u002Fphilkr)\u003C\u002Fli> \u003Cli>[Ishan Misra](https:\u002F\u002Fimisra.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_035daec47c0f.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20077-9_21) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002FDetic?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FDetic) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.02605)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flvis-dataset\u002Flvis-api)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1QtTW9-ukX2HKZGvt0QvVGqjuqEykoZKI) | 07.06.2022 |\n| SimCTG | Contrastive training objective to calibrate the model's representation space, and a decoding method -- contrastive search -- to encourage diversity while maintaining coherence in the generated text | \u003Cul>\u003Cli>[Yixuan Su](https:\u002F\u002Fyxuansu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tian Lan](https:\u002F\u002Fgithub.com\u002FgmftbyGMFTBY)\u003C\u002Fli> \u003Cli>[Yan Wang](https:\u002F\u002Flibertywing.github.io\u002Fyanwang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Dani Yogatama](https:\u002F\u002Fdyogatama.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lingpeng Kong](https:\u002F\u002Fikekonglp.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nigel Collier](https:\u002F\u002Fgithub.com\u002Fnhcollier)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyxuansu\u002FSimCTG?style=social)](https:\u002F\u002Fgithub.com\u002Fyxuansu\u002FSimCTG) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2202.06417), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.14140)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyxuansu\u002FContrastive_Search_versus_Contrastive_Decoding), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyxuansu\u002FContrastive_Search_Is_What_You_Need)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fintroducing-csearch), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fjoaogante\u002Fcontrastive_search_generation), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Fopt)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002F871cae8f599cb8bbfcb0f58fe1af95ad-Abstract-Conference.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fsimctg\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1ImvR-ldHf9rJyFzOCMJ_zjAGK8n1iTW7) | 04.06.2022 |\n| T0 | Multitask Prompted Training Enables Zero-Shot Task Generalization | \u003Cul>\u003Cli>[Victor Sanh](https:\u002F\u002Fgithub.com\u002FVictorSanh)\u003C\u002Fli> \u003Cli>[Albert Webson](https:\u002F\u002Frepresentation.ai\u002F)\u003C\u002Fli> \u003Cli>[Colin Raffel](https:\u002F\u002Fcolinraffel.com\u002F\u002F)\u003C\u002Fli> \u003Cli>[Stephen Bach](http:\u002F\u002Fcs.brown.edu\u002Fpeople\u002Fsbach\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lintang Sutawika](https:\u002F\u002Fgithub.com\u002Flintangsutawika)\u003C\u002Fli> \u003Cli>[Zaid Alyafeai](https:\u002F\u002Fgithub.com\u002Fzaidalyafeai)\u003C\u002Fli> \u003Cli>[Antoine Chaffin](https:\u002F\u002Fantoine.chaffin.fr\u002F)\u003C\u002Fli> \u003Cli>[Arnaud Stiegler](https:\u002F\u002Fgithub.com\u002Farnaudstiegler)\u003C\u002Fli> \u003Cli>[Teven Scao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ik0_vxsAAAAJ)\u003C\u002Fli> \u003Cli>[Arun Raja](https:\u002F\u002Fwww.arunraja.dev\u002F)\u003C\u002Fli> \u003Cli>[Manan Dey](https:\u002F\u002Fgithub.com\u002Fmanandey)\u003C\u002Fli> \u003Cli>[M Saiful Bari](https:\u002F\u002Fsbmaruf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Canwen Xu](https:\u002F\u002Fwww.canwenxu.net\u002F)\u003C\u002Fli> \u003Cli>[Urmish Thakker](https:\u002F\u002Fgithub.com\u002FUrmish)\u003C\u002Fli> \u003Cli>[Shanya Sharma](https:\u002F\u002Fshanyas10.github.io\u002F)\u003C\u002Fli> \u003Cli>[Eliza Szczechla](https:\u002F\u002Felsanns.github.io\u002F)\u003C\u002Fli> \u003Cli>[Taewoon Kim](https:\u002F\u002Ftae898.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gunjan Chhablani](https:\u002F\u002Fgchhablani.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nihal Nayak](https:\u002F\u002Fnihalnayak.github.io\u002F)\u003C\u002Fli> \u003Cli>[Debajyoti Datta](http:\u002F\u002Fdebajyotidatta.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jonathan Chang](https:\u002F\u002Fgithub.com\u002Fcccntu\u002F)\u003C\u002Fli> \u003Cli>[Mike Tian-Jian Jiang](https:\u002F\u002Fgithub.com\u002Ftianjianjiang)\u003C\u002Fli> \u003Cli>[Matteo Manica](https:\u002F\u002Fgithub.com\u002Fdrugilsberg)\u003C\u002Fli> \u003Cli>[Sheng Shen](https:\u002F\u002Fsincerass.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zheng Xin Yong](https:\u002F\u002Fyongzx.github.io\u002F)\u003C\u002Fli> \u003Cli>[Harshit Pandey](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=BPIs78gAAAAJ)\u003C\u002Fli> \u003Cli>[Rachel Bawden](https:\u002F\u002Frbawden.github.io\u002F)\u003C\u002Fli> \u003Cli>[Trishala Neeraj](https:\u002F\u002Fgithub.com\u002Ftrishalaneeraj)\u003C\u002Fli> \u003Cli>[Jos Rozen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=OxEDKogAAAAJ)\u003C\u002Fli> \u003Cli>[Abheesht Sharma](https:\u002F\u002Fgithub.com\u002Fabheesht-sharma)\u003C\u002Fli> \u003Cli>[Andrea Santilli](https:\u002F\u002Fteelinsan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Thibault Fevry](http:\u002F\u002Fthibaultfevry.com\u002F)\u003C\u002Fli> \u003Cli>[Jason Alan Fries](https:\u002F\u002Fweb.stanford.edu\u002F~jfries\u002F)\u003C\u002Fli> \u003Cli>[Ryan Teehan](https:\u002F\u002Fgithub.com\u002Frteehas)\u003C\u002Fli> \u003Cli>[Stella Biderman](https:\u002F\u002Fwww.stellabiderman.com\u002F)\u003C\u002Fli> \u003Cli>[Leo Gao](https:\u002F\u002Fgithub.com\u002Fleogao2)\u003C\u002Fli> \u003Cli>[Tali Bers](https:\u002F\u002Fgithub.com\u002Ftbers-coursera)\u003C\u002Fli> \u003Cli>[Thomas Wolf](https:\u002F\u002Fthomwolf.io\u002F)\u003C\u002Fli> \u003Cli>[Alexander Rush](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=LIjnUGgAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fbigscience-workshop\u002Fpromptsource?style=social)](https:\u002F\u002Fgithub.com\u002Fbigscience-workshop\u002Fpromptsource) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2110.08207)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FiJ0IVZgGjTM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYToXXfrIu6w)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1xx7SgdLaAu23YFBirXmaQViDr8caowX_) | 29.05.2022 |\n| AvatarCLIP | A zero-shot text-driven framework for 3D avatar generation and animation | \u003Cul>\u003Cli>[Fangzhou Hong](https:\u002F\u002Fhongfz16.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mingyuan Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2QLD4fAAAAAJ)\u003C\u002Fli> \u003Cli>[Liang Pan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=lSDISOcAAAAJ)\u003C\u002Fli> \u003Cli>[Zhongang Cai](https:\u002F\u002Fcaizhongang.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lei Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jZH2IPYAAAAJ)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0d0b7192547c.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3528223.3530094) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhongfz16\u002FAvatarCLIP?style=social)](https:\u002F\u002Fgithub.com\u002Fhongfz16\u002FAvatarCLIP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.08535), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.01455), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.03221), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.05139), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.13333)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fwww.di.ens.fr\u002Fwillow\u002Fresearch\u002Fsurreal\u002Fdata\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdaniilidis-group\u002Fneural_renderer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FGuyTevet\u002FMotionCLIP), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTotoro97\u002FNeuS), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnghorbani\u002Fhuman_body_prior)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fhongfz16.github.io\u002Fprojects\u002FAvatarCLIP.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-l2ZMeoASGY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1dfaecX7xF3nP6fyXc8XBljV5QY1lc1TR) | 15.05.2022 |\n| Text2Mesh | Text-Driven Neural Stylization for Meshes | \u003Cul>\u003Cli>[Oscar Michel](https:\u002F\u002Fojmichel.github.io\u002F)\u003C\u002Fli> \u003Cli>[Roi Bar-On](https:\u002F\u002Fgithub.com\u002Froibaron)\u003C\u002Fli> \u003Cli>[Richard Liu](https:\u002F\u002Fgithub.com\u002Ffactoryofthesun)\u003C\u002Fli> \u003Cli>[Sagie Benaim](https:\u002F\u002Fsagiebenaim.github.io\u002F)\u003C\u002Fli> \u003Cli>[Rana Hanocka](http:\u002F\u002Fpeople.cs.uchicago.edu\u002F~ranahanocka\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fthreedle\u002Ftext2mesh?style=social)](https:\u002F\u002Fgithub.com\u002Fthreedle\u002Ftext2mesh) \u003Cul>\u003Cli>[CLIP](https:\u002F\u002Fopenai.com\u002Fblog\u002Fclip\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.03221)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fcode\u002Fneverix\u002Ftext2mesh\u002Fnotebook)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fthreedle.github.io\u002Ftext2mesh\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fthreedle\u002Ftext2mesh\u002Fblob\u002Fmaster\u002Fcolab_demo.ipynb) | 14.05.2022 |\n| T5 | Text-To-Text Transfer Transformer | \u003Cul>\u003Cli>[Colin Raffel](https:\u002F\u002Fcolinraffel.com\u002F)\u003C\u002Fli> \u003Cli>[Noam Shazeer](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=wsGvgA8AAAAJ)\u003C\u002Fli> \u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Katherine Lee](https:\u002F\u002Fgithub.com\u002Fkatelee168)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Sharan Narang](https:\u002F\u002Fgithub.com\u002Fsharannarang)\u003C\u002Fli> \u003Cli>[Michael Matena](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=rN_9vroAAAAJ)\u003C\u002Fli> \u003Cli>[Yanqi Zhou](https:\u002F\u002Fzhouyanqi.github.io)\u003C\u002Fli> \u003Cli>[Wei Li](https:\u002F\u002Fresearch.google\u002Fpeople\u002F106528\u002F)\u003C\u002Fli> \u003Cli>[Peter J. Liu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=1EPxhywAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Ftext-to-text-transfer-transformer?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Ftext-to-text-transfer-transformer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.10683)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fmesh\u002Ftree\u002Fmaster\u002Fmesh_tensorflow\u002Ftransformer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fdatasets)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Ftext-to-text-transfer-transformer\u002Fblob\u002Fmain\u002Fnotebooks\u002Ft5-trivia.ipynb) | 11.05.2022 |\n| XLS-R | Self-supervised Cross-lingual Speech Representation Learning at Scale | \u003Cul>\u003Cli>[Arun Babu](https:\u002F\u002Fgithub.com\u002Farbabu123)\u003C\u002Fli> \u003Cli>[Changhan Wang](https:\u002F\u002Fwww.changhan.me\u002F)\u003C\u002Fli> \u003Cli>[Andros Tjandra](https:\u002F\u002Fgithub.com\u002Fandrostj)\u003C\u002Fli> \u003Cli>[Kushal Lakhotia](https:\u002F\u002Fabout.me\u002Fhikushalhere)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Qiantong Xu](https:\u002F\u002Fgithub.com\u002Fxuqiantong)\u003C\u002Fli> \u003Cli>[Naman Goyal](https:\u002F\u002Fgithub.com\u002Fngoyal2707)\u003C\u002Fli> \u003Cli>[Kritika Singh](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Ltk3SykAAAAJ)\u003C\u002Fli> \u003Cli>[Patrick von Platen](https:\u002F\u002Fgithub.com\u002Fpatrickvonplaten)\u003C\u002Fli> \u003Cli>[Yatharth Saraf](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=KJTtNJwAAAAJ)\u003C\u002Fli> \u003Cli>[Juan Pino](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=weU_-4IAAAAJ)\u003C\u002Fli> \u003Cli>[Alexei Baevski](https:\u002F\u002Fgithub.com\u002Falexeib)\u003C\u002Fli> \u003Cli>[Alexis Conneau](https:\u002F\u002Fgithub.com\u002Faconneau)\u003C\u002Fli> \u003Cli>[Michael Auli](https:\u002F\u002Fgithub.com\u002Fmichaelauli)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Ffairseq?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffairseq\u002Fblob\u002Fmain\u002Fexamples\u002Fwav2vec\u002Fxlsr\u002FREADME.md) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.09296)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Ffine-tune-xlsr-wav2vec2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffairscale)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fpatrickvonplaten\u002Fnotebooks\u002Fblob\u002Fmaster\u002FFine_Tune_XLS_R_on_Common_Voice.ipynb) | 10.05.2022 |\n| MAGIC | Training-free framework, iMAge-Guided text generatIon with CLIP, for plugging in visual controls in the generation process and enabling LMs to perform multimodal tasks in a zero-shot manner | \u003Cul>\u003Cli>[Yixuan Su](https:\u002F\u002Fyxuansu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tian Lan](https:\u002F\u002Fgithub.com\u002FgmftbyGMFTBY)\u003C\u002Fli> \u003Cli>[Yahui Liu](https:\u002F\u002Fyhlleo.github.io\u002F)\u003C\u002Fli> \u003Cli>[Fangyu Liu](https:\u002F\u002Ffangyuliu.me\u002Fabout)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dani Yogatama](https:\u002F\u002Fdyogatama.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yan Wang](https:\u002F\u002Flibertywing.github.io\u002Fyanwang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Lingpeng Kong](https:\u002F\u002Fwww.cs.cmu.edu\u002F~lingpenk\u002F)\u003C\u002Fli> \u003Cli>[Nigel Collier](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fnhcollier\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyxuansu\u002Fmagic?style=social)](https:\u002F\u002Fgithub.com\u002Fyxuansu\u002Fmagic) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.02655)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1NDVkKpanbsaUwecHoRp_2kIpMztOFW25) | 02.05.2022 |\n| DiffCSE | Unsupervised contrastive learning framework for learning sentence embeddings | \u003Cul>\u003Cli>[Yung-Sung Chuang](https:\u002F\u002Fpeople.csail.mit.edu\u002Fyungsung\u002F)\u003C\u002Fli> \u003Cli>[Rumen Dangovski](http:\u002F\u002Fsuper-ms.mit.edu\u002Frumen.html)\u003C\u002Fli> \u003Cli>[Hongyin Luo](https:\u002F\u002Fluohongyin.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yang Zhang](https:\u002F\u002Fmitibmwatsonailab.mit.edu\u002Fpeople\u002Fyang-zhang\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Shiyu Chang](https:\u002F\u002Fcode-terminator.github.io\u002F)\u003C\u002Fli> \u003Cli>[Marin Soljačić](http:\u002F\u002Fwww.mit.edu\u002F~soljacic\u002Fmarin.html)\u003C\u002Fli> \u003Cli>[Shang-Wen Li](https:\u002F\u002Fswdanielli.github.io\u002F)\u003C\u002Fli> \u003Cli>[Scott Wen-tau Yih](https:\u002F\u002Fscottyih.org\u002F)\u003C\u002Fli> \u003Cli>[Yoon Kim](https:\u002F\u002Fpeople.csail.mit.edu\u002Fyoonkim\u002F)\u003C\u002Fli> \u003Cli>[James Glass](http:\u002F\u002Fgroups.csail.mit.edu\u002Fsls\u002Fpeople\u002Fglass.shtml)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fvoidism\u002Fdiffcse?style=social)](https:\u002F\u002Fgithub.com\u002Fvoidism\u002Fdiffcse) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.10298), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.08821), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.00899)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fprinceton-nlp\u002FSimCSE)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fvoidism)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002FYungSungChuang\u002Fstatus\u002F1517518077902000129)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fvoidism\u002FDiffCSE\u002Fblob\u002Fmaster\u002Fdiffcse_evaluation.ipynb) | 24.04.2022 |\n| ViDT+ | An Extendable, Efficient and Effective Transformer-based Object Detector | \u003Cul>\u003Cli>[Hwanjun Song](https:\u002F\u002Fsonghwanjun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Deqing Sun](https:\u002F\u002Fdeqings.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sanghyuk Chun](https:\u002F\u002Fsanghyukchun.github.io\u002Fhome\u002F)\u003C\u002Fli> \u003Cli>[Varun Jampani](https:\u002F\u002Fvarunjampani.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dongyoon Han](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fdyhan0920\u002F)\u003C\u002Fli> \u003Cli>[Byeongho Heo](https:\u002F\u002Fsites.google.com\u002Fview\u002Fbyeongho-heo\u002Fhome)\u003C\u002Fli> \u003Cli>[Wonjae Kim](https:\u002F\u002Fwonjae.kim\u002F)\u003C\u002Fli> \u003Cli>[Ming-Hsuan Yang](http:\u002F\u002Ffaculty.ucmerced.edu\u002Fmhyang\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fnaver-ai\u002Fvidt?style=social)](https:\u002F\u002Fgithub.com\u002Fnaver-ai\u002Fvidt\u002Ftree\u002Fvidt-plus) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.07962), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2110.03921)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffundamentalvision\u002FDeformable-DETR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEherSenaw\u002FViDT_colab)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FEherSenaw\u002FViDT_colab\u002Fblob\u002Fmain\u002Fvidt_colab.ipynb) | 20.04.2022 |\n| BasicVSR++ | Redesign BasicVSR by proposing second-order grid propagation and flow-guided deformable alignment | \u003Cul>\u003Cli>[Kelvin Chan](https:\u002F\u002Fckkelvinchan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shangchen Zhou](https:\u002F\u002Fshangchenzhou.com\u002F)\u003C\u002Fli> \u003Cli>[Xiangyu Xu](https:\u002F\u002Fxuxy09.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c2a923ac3976.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00588) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fckkelvinchan\u002FBasicVSR_PlusPlus?style=social)](https:\u002F\u002Fgithub.com\u002Fckkelvinchan\u002FBasicVSR_PlusPlus) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.13371)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fckkelvinchan\u002FBasicVSR-IconVSR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fckkelvinchan\u002Foffset-fidelity-loss)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fckkelvinchan.github.io\u002Fprojects\u002FBasicVSR++\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FiIDml09CUc4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1I0kZMM0DQyb4ueHZw5si8fMnRCJ_eUX3) | 18.04.2022 |\n| NAFNet | Nonlinear Activation Free Network for Image Restoration | \u003Cul>\u003Cli>[Liangyu Chen](https:\u002F\u002Fgithub.com\u002Fmayorx)\u003C\u002Fli> \u003Cli>[Xiaojie Chu](https:\u002F\u002Fgithub.com\u002Fchuxiaojie)\u003C\u002Fli> \u003Cli>[Xiangyu Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=yuB-cfoAAAAJ)\u003C\u002Fli> \u003Cli>[Jian Sun](http:\u002F\u002Fwww.jiansun.org\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_827efba6b6c2.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20071-7_2) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmegvii-research\u002FNAFNet?style=social)](https:\u002F\u002Fgithub.com\u002Fmegvii-research\u002FNAFNet) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.04676), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.08714)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fimage-deblurring-on-gopro?p=simple-baselines-for-image-restoration), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fimage-denoising-on-sidd?p=simple-baselines-for-image-restoration)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1dkO5AyktmBoWwxBwoKFUurIDn0m4qDXT) | 15.04.2022 |\n| Panini-Net | GAN Prior based Degradation-Aware Feature Interpolation for Face Restoration | \u003Cul>\u003Cli>[Yinhuai Wang](https:\u002F\u002Fgithub.com\u002Fwyhuai)\u003C\u002Fli> \u003Cli>[Yujie Hu](https:\u002F\u002Fvilla.jianzhang.tech\u002Fpeople\u002Fyujie-hu\u002F)\u003C\u002Fli> \u003Cli>[Jian Zhang](http:\u002F\u002Fjianzhang.tech\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_dc79c78aa8d7.png)](https:\u002F\u002Fdoi.org\u002F10.1609\u002Faaai.v36i3.20159) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjianzhangcs\u002Fpanini?style=social)](https:\u002F\u002Fgithub.com\u002Fjianzhangcs\u002Fpanini) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.08444)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftkarras\u002Fprogressive_growing_of_gans)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGeeveGeorge\u002FPanini-Net-Colab\u002Fblob\u002Fmain\u002FPaniniNet_Working.ipynb) | 13.04.2022 |\n| E2FGVI | An End-to-End framework for Flow-Guided Video Inpainting through elaborately designed three trainable modules, namely, flow completion, feature propagation, and content hallucination modules | \u003Cul>\u003Cli>[Zhen Li](https:\u002F\u002Fpaper99.github.io\u002F)\u003C\u002Fli> \u003Cli>[Cheng-Ze Lu](https:\u002F\u002Fgithub.com\u002FLGYoung)\u003C\u002Fli> \u003Cli>[Jianhua Qin](https:\u002F\u002Fscholar.google.com\u002Fcitations?&user=TAr7TU4AAAAJ)\u003C\u002Fli> \u003Cli>[Chun-Le Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RZLYwR0AAAAJ)\u003C\u002Fli> \u003Cli>[Ming-Ming Cheng](https:\u002F\u002Fmmcheng.net\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9e584c39a658.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01704) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FMCG-NKU\u002FE2FGVI?style=social)](https:\u002F\u002Fgithub.com\u002FMCG-NKU\u002FE2FGVI) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.02663)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcompetitions.codalab.org\u002Fcompetitions\u002F19544#participate-get-data), [data](https:\u002F\u002Fdata.vision.ee.ethz.ch\u002Fcsergi\u002Fshare\u002Fdavis\u002FDAVIS-2017-trainval-480p.zip)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fresearchmm\u002FSTTN), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FFocal-Transformer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fruiliu-ai\u002FFuseFormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fphoenix104104\u002Ffast_blind_video_consistency#evaluation)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fmlearning-ai\u002Fend-to-end-framework-for-flow-guided-video-inpainting-c5e2d8b61d20)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FN--qC3T2wc4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3eH3Fm6gOFk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F12rwY2gtG8jVWlNx9pjmmM8uGmh5ue18G) | 06.04.2022 |\n| LDM | High-Resolution Image Synthesis with Latent Diffusion Models | \u003Cul>\u003Cli>[Robin Rombach](https:\u002F\u002Fgithub.com\u002Frromb)\u003C\u002Fli> \u003Cli>[Andreas Blattmann](https:\u002F\u002Fgithub.com\u002Fablattmann)\u003C\u002Fli> \u003Cli>[Dominik Lorenz](https:\u002F\u002Fgithub.com\u002Fqp-qp)\u003C\u002Fli> \u003Cli>[Patrick Esser](https:\u002F\u002Fgithub.com\u002Fpesser)\u003C\u002Fli> \u003Cli>[Björn Ommer](https:\u002F\u002Fommer-lab.com\u002Fpeople\u002Fommer\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_7578c7a947c2.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01042) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCompVis\u002Flatent-diffusion?style=social)](https:\u002F\u002Fgithub.com\u002FCompVis\u002Flatent-diffusion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10752), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2202.09778), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.02114)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffyu\u002Flsun), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fguided-diffusion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flucidrains\u002Fdenoising-diffusion-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flucidrains\u002Fx-transformers)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fmultimodalart\u002Flatentdiffusion)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FCompVis\u002Flatent-diffusion\u002Fblob\u002Fmaster\u002Fscripts\u002Flatent_imagenet_diffusion.ipynb) | 04.04.2022 |\n| GP-UNIT | Novel framework, Generative Prior-guided UNsupervised Image-to-image Translation, to improve the overall quality and applicability of the translation algorithm | \u003Cul>\u003Cli>[Shuai Yang](https:\u002F\u002Fwilliamyang1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liming Jiang](https:\u002F\u002Fliming-jiang.com\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwilliamyang1991\u002FGP-UNIT?style=social)](https:\u002F\u002Fgithub.com\u002Fwilliamyang1991\u002FGP-UNIT) \u003Cul>\u003Cli>[ImageNet](https:\u002F\u002Fimage-net.org\u002Fdownload.php)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.03641)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fclovaai\u002Fstargan-v2#datasets-and-pre-trained-networks), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fswitchablenorms\u002FCelebAMask-HQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fmetfaces-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTreB1eN\u002FInsightFace_Pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002FSPADE), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnvlabs\u002Fimaginaire), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01779)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fproject\u002Fgpunit\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdDApWs_oDrM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fwilliamyang1991\u002FGP-UNIT\u002Fblob\u002Fmain\u002Fnotebooks\u002Finference_playground.ipynb) | 02.04.2022 |\n| DualStyleGAN | More challenging exemplar-based high-resolution portrait style transfer by introducing a novel DualStyleGAN with flexible control of dual styles of the original face domain and the extended artistic portrait domain | \u003Cul>\u003Cli>[Shuai Yang](https:\u002F\u002Fwilliamyang1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liming Jiang](https:\u002F\u002Fliming-jiang.com\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0239aadbb33a.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00754) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwilliamyang1991\u002FDualStyleGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fwilliamyang1991\u002FDualStyleGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.13248)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcs.nju.edu.cn\u002Frl\u002FWebCaricature.htm), [data](https:\u002F\u002Fwww.gwern.net\u002FCrops#danbooru2019-portraits)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flowfuel\u002Fprogrock-stable), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTreB1eN\u002FInsightFace_Pytorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FGradio-Blocks\u002FDualStyleGAN), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fhysts\u002FDualStyleGAN)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fproject\u002Fdualstylegan\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FscZTu77jixI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fwilliamyang1991\u002FDualStyleGAN\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 24.03.2022 |\n| CLIPasso | Semantically-Aware Object Sketching | \u003Cul>\u003Cli>[Yael Vinker](https:\u002F\u002Fyaelvi116.wixsite.com\u002Fmysite)\u003C\u002Fli> \u003Cli>[Ehsan Pajouheshgar](https:\u002F\u002Fpajouheshgar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jessica Y. Bo](https:\u002F\u002Fjessica-bo.github.io\u002F)\u003C\u002Fli> \u003Cli>[Roman Bachmann](https:\u002F\u002Froman-bachmann.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Amit Bermano](https:\u002F\u002Fwww.cs.tau.ac.il\u002F~amberman\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli> \u003Cli>[Amir Zamir](https:\u002F\u002Fvilab.epfl.ch\u002Fzamir\u002F)\u003C\u002Fli> \u003Cli>[Ariel Shamir](https:\u002F\u002Ffaculty.runi.ac.il\u002Farik\u002Fsite\u002Findex.asp)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyael-vinker\u002FCLIPasso?style=social)](https:\u002F\u002Fgithub.com\u002Fyael-vinker\u002FCLIPasso) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2202.05822), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.14843)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Freplicate.com\u002Fyael-vinker\u002Fclipasso)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FBachiLi\u002Fdiffvg)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fclipasso.github.io\u002Fclipasso\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyael-vinker\u002FCLIPasso\u002Fblob\u002Fmain\u002FCLIPasso.ipynb) | 21.03.2022 |\n| StyleSDF | A high resolution, 3D-consistent image and shape generation technique | \u003Cul>\u003Cli>[Roy Or-El](https:\u002F\u002Fhomes.cs.washington.edu\u002F~royorel\u002F)\u003C\u002Fli> \u003Cli>[Xuan Luo](https:\u002F\u002Froxanneluo.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mengyi Shan](https:\u002F\u002Fshanmy.github.io\u002F)\u003C\u002Fli> \u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jeong Joon Park](https:\u002F\u002Fjjparkcv.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ira Kemelmacher-Shlizerman](https:\u002F\u002Fwww.irakemelmacher.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_914392516a6b.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01314) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Froyorel\u002FStyleSDF?style=social)](https:\u002F\u002Fgithub.com\u002Froyorel\u002FStyleSDF) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.11427)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyenchenlin\u002Fnerf-pytorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FSerdarHelli\u002FStyleSDF-3D)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fstylesdf.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Froyorel\u002FStyleSDF\u002Fblob\u002Fmain\u002FStyleSDF_demo.ipynb) | 05.03.2022 |\n| Disentangled Lifespan Face Synthesis | LFS model is proposed to disentangle the key face characteristics including shape, texture and identity so that the unique shape and texture age transformations can be modeled effectively | \u003Cul>\u003Cli>[Sen He](https:\u002F\u002Fsenhe.github.io\u002F)\u003C\u002Fli> \u003Cli>[Wentong Liao](https:\u002F\u002Fwww.tnt.uni-hannover.de\u002Fen\u002Fstaff\u002Fliao\u002F)\u003C\u002Fli> \u003Cli>[Michael Yang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fmichaelyingyang\u002F)\u003C\u002Fli> \u003Cli>[Yi-Zhe Song](http:\u002F\u002Fpersonal.ee.surrey.ac.uk\u002FPersonal\u002FY.Song\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Bodo Rosenhahn](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=qq3TxtcAAAAJ)\u003C\u002Fli> \u003Cli>[Tao Xiang](http:\u002F\u002Fpersonal.ee.surrey.ac.uk\u002FPersonal\u002FT.Xiang\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FSenHe\u002FDLFS?style=social)](https:\u002F\u002Fgithub.com\u002FSenHe\u002FDLFS) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.02874)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsenhe.github.io\u002Fprojects\u002Ficcv_2021_lifespan_face\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=uklX03ns0m0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1fgVAoxCSaqPkj0rUK4RmBh7GTQRqLNpE) | 22.02.2022 |\n| ClipCap | CLIP Prefix for Image Captioning | \u003Cul>\u003Cli>[Ron Mokady](https:\u002F\u002Frmokady.github.io\u002F)\u003C\u002Fli> \u003Cli>[Amir Hertz](https:\u002F\u002Fgithub.com\u002Famirhertz)\u003C\u002Fli> \u003Cli>[Amit Bermano](https:\u002F\u002Fwww.cs.tau.ac.il\u002F~amberman\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Frmokady\u002FCLIP_prefix_caption?style=social)](https:\u002F\u002Fgithub.com\u002Frmokady\u002FCLIP_prefix_caption) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.09734)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcocodataset.org\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002FCLIP_prefix_captioning)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@uppalamukesh\u002Fclipcap-clip-prefix-for-image-captioning-3970c73573bc)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVQDrmuccWDo)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Frmokady\u002FCLIP_prefix_caption\u002Fblob\u002Fmain\u002Fnotebooks\u002Fclip_prefix_captioning_inference.ipynb#scrollTo=glBzYsgIwhwF) | 15.02.2022 |\n| ROMP | Monocular, One-stage, Regression of Multiple 3D People | \u003Cul>\u003Cli>[Yu Sun](https:\u002F\u002Fwww.yusun.work\u002F)\u003C\u002Fli> \u003Cli>[Qian Bao](https:\u002F\u002Fgithub.com\u002Ffor-code0216)\u003C\u002Fli> \u003Cli>[Wu Liu](https:\u002F\u002Ffaculty.ustc.edu.cn\u002Fliuwu)\u003C\u002Fli> \u003Cli>[Yili Fu](https:\u002F\u002Fieeexplore.ieee.org\u002Fauthor\u002F37286601800)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli> \u003Cli>[Tao Mei](https:\u002F\u002Ftaomei.me\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fac7d2fbad55.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.01099) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FArthur151\u002FROMP?style=social)](https:\u002F\u002Fgithub.com\u002FArthur151\u002FROMP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.12272), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.08274), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F2306.02850)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FArthur151\u002FRelative_Human), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FArthur151\u002FDynaCam), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyanchxx\u002FMoPA)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhunBPJxnyBU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQ62fj_6AxRI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fl8aLHDXWQRw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1oz9E6uIbj4udOPZvA1Zi9pFx0SWH_UXg) | 11.02.2022 |\n| Mask2Former | Masked-attention Mask Transformer for Universal Image Segmentation | \u003Cul>\u003Cli>[Bowen Cheng](https:\u002F\u002Fbowenc0221.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ishan Misra](https:\u002F\u002Fimisra.github.io\u002F)\u003C\u002Fli> \u003Cli>[Alexander Schwing](https:\u002F\u002Falexander-schwing.de\u002F)\u003C\u002Fli> \u003Cli>[Alexander Kirillov](https:\u002F\u002Falexander-kirillov.github.io\u002F)\u003C\u002Fli> \u003Cli>[Rohit Girdhar](https:\u002F\u002Frohitgirdhar.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_7b9e52c86acd.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00135) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002FMask2Former?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FMask2Former) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.01527), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10764)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Freplicate.com\u002Ffacebookresearch\u002Fmask2former)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FMaskFormer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002FMask2Former)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fbowenc0221.github.io\u002Fmask2former\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1uIWE5KbGFSjrxey2aRd5pWkKNY1_SaNq) | 09.02.2022 |\n| BertViz | Tool that visualizes attention at multiple scales, each of which provides a unique perspective on the attention mechanism | [Jesse Vig](https:\u002F\u002Fjessevig.com\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_28aa197db431.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002FP19-3007) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjessevig\u002Fbertviz?style=social)](https:\u002F\u002Fgithub.com\u002Fjessevig\u002Fbertviz) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.05714), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.11218), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.10186), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.04626), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.05607)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@GaryFr0sty\u002Fvisualize-attention-scores-of-llms-with-bertviz-3deb94b455b3)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fbertviz\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FumErGRrfSk4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1hXIQ77A4TYS4y3UthWF-Ci7V7vVUoxmQ) | 05.02.2022 |\n| JoJoGAN | One Shot Face Stylization | \u003Cul>\u003Cli>[Min Jin Chong](https:\u002F\u002Fmchong6.github.io\u002F)\u003C\u002Fli> \u003Cli>[David Forsyth](http:\u002F\u002Fluthuli.cs.uiuc.edu\u002F~daf\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_b840583229af.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-19787-1_8) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmchong6\u002FJoJoGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fmchong6\u002FJoJoGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.11641)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Freplicate\u002Fcog)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmchong6\u002FJoJoGAN\u002Fblob\u002Fmaster\u002Fstylize.ipynb) | 02.02.2022 |\n| Pose with Style | Detail-Preserving Pose-Guided Image Synthesis with Conditional StyleGAN | \u003Cul>\u003Cli>[Badour AlBahar](https:\u002F\u002Fbadouralbahar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jingwan Lu](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Fjingwan-lu\u002F)\u003C\u002Fli> \u003Cli>[Jimei Yang](https:\u002F\u002Fgithub.com\u002Fjimeiyang)\u003C\u002Fli> \u003Cli>[Zhixin Shu](https:\u002F\u002Fzhixinshu.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli> \u003Cli>[Jia-Bin Huang](https:\u002F\u002Fjbhuang0604.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FBadourAlBahar\u002Fpose-with-style?style=social)](https:\u002F\u002Fgithub.com\u002FBadourAlBahar\u002Fpose-with-style) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.06166)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpose-with-style.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fd_ETeAVLilw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftg-bomze\u002Fcollection-of-notebooks\u002Fblob\u002Fmaster\u002FHomeStylist.ipynb) | 19.01.2022 |\n| ConvNeXt | A pure ConvNet model constructed entirely from standard ConvNet modules | \u003Cul>\u003Cli>[Zhuang Liu](https:\u002F\u002Fliuzhuang13.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hanzi Mao](https:\u002F\u002Fhanzimao.me\u002F)\u003C\u002Fli> \u003Cli>[Chao-Yuan Wu](https:\u002F\u002Fchaoyuan.org\u002F)\u003C\u002Fli> \u003Cli>[Christoph Feichtenhofer](https:\u002F\u002Ffeichtenhofer.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Trevor Darrell](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~trevor\u002F)\u003C\u002Fli> \u003Cli>[Saining Xie](https:\u002F\u002Fwww.sainingxie.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_5f262d9dec0d.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01167) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002FConvNeXt?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FConvNeXt) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.03545)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frwightman\u002Fpytorch-image-models), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdeit)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002Fconvnext)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQzCjXqFnWPE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FidiIllIQOfU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQqejV0LNDHA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1CBYTIZ4tBMsVL5cqu9N_-Q3TBprqsfEO) | 19.01.2022 |\n| diffsort | Differentiable Sorting Networks | \u003Cul>\u003Cli>[Felix Petersen](http:\u002F\u002Fpetersen.ai\u002F)\u003C\u002Fli> \u003Cli>[Christian Borgelt](https:\u002F\u002Fborgelt.net\u002F)\u003C\u002Fli> \u003Cli>[Hilde Kuehne](https:\u002F\u002Fhildekuehne.github.io\u002F)\u003C\u002Fli> \u003Cli>[Oliver Deussen](https:\u002F\u002Fwww.cgmi.uni-konstanz.de\u002Fpersonen\u002Fprof-dr-oliver-deussen\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FFelix-Petersen\u002Fdiffsort?style=social)](https:\u002F\u002Fgithub.com\u002FFelix-Petersen\u002Fdiffsort) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.04019), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.09630)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRl-sFaE1z4M)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1q0TZFFYB9FlOJYWKt0_7ZaXQT190anhm) | 17.01.2022 |\n| Taming Transformers for High-Resolution Image Synthesis | We combine the efficiancy of convolutional approaches with the expressivity of transformers by introducing a convolutional VQGAN, which learns a codebook of context-rich visual parts, whose composition is modeled with an autoregressive transformer | \u003Cul>\u003Cli>[Patrick Esser](https:\u002F\u002Fgithub.com\u002Fpesser)\u003C\u002Fli> \u003Cli>[Robin Rombach](https:\u002F\u002Fgithub.com\u002Frromb)\u003C\u002Fli> \u003Cli>[Björn Ommer](https:\u002F\u002Fommer-lab.com\u002Fpeople\u002Fommer\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_df248106cc28.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01268) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCompVis\u002Ftaming-transformers?style=social)](https:\u002F\u002Fgithub.com\u002FCompVis\u002Ftaming-transformers) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.09841)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fcompvis.github.io\u002Ftaming-transformers\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FCompVis\u002Ftaming-transformers\u002Fblob\u002Fmaster\u002Fscripts\u002Ftaming-transformers.ipynb) | 13.01.2022 |\n| GFM | Glance and Focus Matting network, which employs a shared encoder and two separate decoders to learn both tasks in a collaborative manner for end-to-end natural image matting | \u003Cul>\u003Cli>[Jizhizi Li](https:\u002F\u002Fjizhizili.github.io\u002Fhomepage\u002F)\u003C\u002Fli> \u003Cli>[Jing Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=9jH5v74AAAAJ)\u003C\u002Fli> \u003Cli>[Stephen Maybank](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=gpyHJmcAAAAJ)\u003C\u002Fli> \u003Cli>[Dacheng Tao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RwlJNLcAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FJizhiziLi\u002FGFM?style=social)](https:\u002F\u002Fgithub.com\u002FJizhiziLi\u002FGFM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.16188)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJizhiziLi\u002FRIM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJizhiziLi\u002FP3M), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJizhiziLi\u002FAIM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FViTAE-Transformer\u002FP3M-Net), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FjizhiziLi\u002Fmatting-survey)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FFJPm4YQOEyo)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1EaQ5h4u9Q_MmDSFTDmFG0ZOeSsFuRTsJ) | 05.01.2022 |\n| RealBasicVSR | Investigating Tradeoffs in Real-World Video Super-Resolution | \u003Cul>\u003Cli>[Kelvin Chan](https:\u002F\u002Fckkelvinchan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shangchen Zhou](https:\u002F\u002Fshangchenzhou.com\u002F)\u003C\u002Fli> \u003Cli>[Xiangyu Xu](https:\u002F\u002Fxuxy09.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_137a3122300c.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00587) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fckkelvinchan\u002FRealBasicVSR?style=social)](https:\u002F\u002Fgithub.com\u002Fckkelvinchan\u002FRealBasicVSR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.12704)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002FRealBasicVSR)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Ftc8p70\u002Frp_investigating_tradeoffs_in_realworld_video\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1JzWRUR34hpKvtCHm84IGx6nv35LCv20J) | 25.12.2021 |\n| GLIDE | Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models | \u003Cul>\u003Cli>[Alex Nichol](https:\u002F\u002Faqnichol.com\u002F)\u003C\u002Fli> \u003Cli>[Prafulla Dhariwal](https:\u002F\u002Fgithub.com\u002Fprafullasd)\u003C\u002Fli> \u003Cli>[Aditya Ramesh](http:\u002F\u002Fadityaramesh.com\u002F)\u003C\u002Fli> \u003Cli>[Pranav Shyam](https:\u002F\u002Fgithub.com\u002Fpranv)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Pamela Mishkin](https:\u002F\u002Fmanlikemishap.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bob McGrew](https:\u002F\u002Fgithub.com\u002Fbmcgrew)\u003C\u002Fli> \u003Cli>[Ilya Sutskever](http:\u002F\u002Fwww.cs.utoronto.ca\u002F~ilya\u002F)\u003C\u002Fli> \u003Cli>[Mark Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5fU-QMwAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fopenai\u002Fglide-text2im?style=social)](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fglide-text2im) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10741)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FItKi3h7IY2o)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fopenai\u002Fglide-text2im\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finpaint.ipynb) | 22.12.2021 |\n| Nerfies | First method capable of photorealistically reconstructing deformable scenes using photos\u002Fvideos captured casually from mobile phones | \u003Cul>\u003Cli>[Keunhong Park](https:\u002F\u002Fkeunhong.com\u002F)\u003C\u002Fli> \u003Cli>[Utkarsh Sinha](https:\u002F\u002Futkarshsinha.com\u002F)\u003C\u002Fli> \u003Cli>[Jon Barron](https:\u002F\u002Fjonbarron.info\u002F)\u003C\u002Fli> \u003Cli>[Sofien Bouaziz](http:\u002F\u002Fsofienbouaziz.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dan Goldman](https:\u002F\u002Fwww.danbgoldman.com\u002Fhome\u002F)\u003C\u002Fli> \u003Cli>[Steve Seitz](https:\u002F\u002Fwww.smseitz.com\u002F)\u003C\u002Fli> \u003Cli>[Ricardo Martin-Brualla](https:\u002F\u002Fricardomartinbrualla.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_20633cc40616.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00581) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fnerfies?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fnerfies) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.12948)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fgoogle-research\u002Ftree\u002Fmaster\u002Fjaxnerf)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnerfies.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fphotogrammetry\u002Fcomments\u002Fk1i0ct\u002Fdeformable_neural_radiance_fields_nerfies\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FMrKrnHhk8IA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIDMiMKWucaI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fnerfies\u002Fblob\u002Fmain\u002Fnotebooks\u002FNerfies_Capture_Processing.ipynb) | 06.12.2021 |\n| HyperStyle | A hypernetwork that learns to modulate StyleGAN's weights to faithfully express a given image in editable regions of the latent space | \u003Cul>\u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Omer Tov](https:\u002F\u002Fgithub.com\u002Fomertov)\u003C\u002Fli> \u003Cli>[Ron Mokady](https:\u002F\u002Frmokady.github.io\u002F)\u003C\u002Fli> \u003Cli>[Rinon Gal](https:\u002F\u002Frinongal.github.io\u002F)\u003C\u002Fli> \u003Cli>[Amit Bermano](https:\u002F\u002Fwww.cs.tau.ac.il\u002F~amberman\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_27a55ce01ec1.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01796) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyuval-alaluf\u002Fhyperstyle?style=social)](https:\u002F\u002Fgithub.com\u002Fyuval-alaluf\u002Fhyperstyle) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.15666), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.03189), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.09036), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.07727)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fai.stanford.edu\u002F~jkrause\u002Fcars\u002Fcar_dataset.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fclovaai\u002Fstargan-v2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTreB1eN\u002FInsightFace_Pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHuangYG123\u002FCurricularFace), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flessw2020\u002FRanger-Deep-Learning-Optimizer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fpytorch\u002Fvision\u002Fblob\u002Fmain\u002Ftorchvision\u002Fmodels\u002Fresnet.py), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdvschultz\u002Fstylegan2-ada-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyuval-alaluf.github.io\u002Fhyperstyle\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_sbXmLY2jMw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyuval-alaluf\u002Fhyperstyle\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 03.12.2021 |\n| encoder4editing | Designing an Encoder for StyleGAN Image Manipulation | \u003Cul>\u003Cli>[Omer Tov](https:\u002F\u002Fgithub.com\u002Fomertov)\u003C\u002Fli> \u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yotam Nitzan](https:\u002F\u002Fyotamnitzan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1a4c1c558661.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3450626.3459838) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fomertov\u002Fencoder4editing?style=social)](https:\u002F\u002Fgithub.com\u002Fomertov\u002Fencoder4editing) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.02766)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Feladrich\u002Fpixel2style2pixel)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fomertov\u002Fencoder4editing\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 02.12.2021 |\n| StyleCariGAN | Caricature Generation via StyleGAN Feature Map Modulation | \u003Cul>\u003Cli>[Wonjong Jang](https:\u002F\u002Fwonjongg.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gwangjin Ju](https:\u002F\u002Fgithub.com\u002Fjugwangjin)\u003C\u002Fli> \u003Cli>[Yucheol Jung](https:\u002F\u002Fycjung.info\u002F)\u003C\u002Fli> \u003Cli>[Jiaolong Yang](https:\u002F\u002Fjlyang.org\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xin Tong](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fxtong\u002F)\u003C\u002Fli> \u003Cli>[Seungyong Lee](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=yGPH-nAAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9efa6c5182cc.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3450626.3459860) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwonjongg\u002FStyleCariGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fwonjongg\u002FStyleCariGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.04331)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwonjongg.github.io\u002FStyleCariGAN\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=kpHbGOlI-BU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1HDRQGm7pvC9mAb6Lktoft_SmY9sCq_Qg) | 30.11.2021 |\n| CartoonGAN | The implementation of the cartoon GAN model with PyTorch | [Tobias Sunderdiek](https:\u002F\u002Fgithub.com\u002FTobiasSunderdiek) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_f01de748abab.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR.2018.00986) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Falamson\u002Fsafebooru)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftobiassunderdiek.github.io\u002Fcartoon-gan\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FTobiasSunderdiek\u002Fcartoon-gan\u002Fblob\u002Fmaster\u002FCartoonGAN.ipynb) | 24.11.2021 |\n| SimSwap | An efficient framework, called Simple Swap, aiming for generalized and high fidelity face swapping | \u003Cul>\u003Cli>[Xuanhong Chen](https:\u002F\u002Fgithub.com\u002Fneuralchen)\u003C\u002Fli> \u003Cli>[Bingbing Ni](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=eUbmKwYAAAAJ)\u003C\u002Fli> \u003Cli>[Yanhao Ge](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=h6tuBAcAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_125d6c68e8af.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3394171.3413630) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fneuralchen\u002FSimSwap?style=social)](https:\u002F\u002Fgithub.com\u002Fneuralchen\u002FSimSwap) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.06340)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fneuralchen\u002FSimSwap\u002Fblob\u002Fmaster\u002FSimSwap%20colab.ipynb) | 24.11.2021 |\n| RVM | Robust High-Resolution Video Matting with Temporal Guidance | \u003Cul>\u003Cli>[Shanchuan Lin](https:\u002F\u002Fgithub.com\u002FPeterL1n)\u003C\u002Fli> \u003Cli>[Linjie Yang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Flinjieyang89\u002F)\u003C\u002Fli> \u003Cli>[Imran Saleemi](http:\u002F\u002Fwww.cs.ucf.edu\u002F~imran\u002F)\u003C\u002Fli> \u003Cli>[Soumyadip Sengupta](https:\u002F\u002Fhomes.cs.washington.edu\u002F~soumya91\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a1152adfc4dc.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FWACV51458.2022.00319) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPeterL1n\u002FRobustVideoMatting?style=social)](https:\u002F\u002Fgithub.com\u002FPeterL1n\u002FRobustVideoMatting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F2108.11515)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FVideoProcessingFramework), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFeiGeChuanShu\u002Fncnn_Android_RobustVideoMatting)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpeterl1n.github.io\u002FRobustVideoMatting)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJvzltozpbpk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAy-mGCEYEzM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F10z-pNKRnVNsp0Lq9tH1J_XPZ7CBC_uHm) | 24.11.2021 |\n| RVM | Robust, real-time, high-resolution human video matting method that achieves new state-of-the-art performance | \u003Cul>\u003Cli>[Shanchuan Lin](https:\u002F\u002Fgithub.com\u002FPeterL1n)\u003C\u002Fli> \u003Cli>[Linjie Yang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Flinjieyang89)\u003C\u002Fli> \u003Cli>[Imran Saleemi](https:\u002F\u002Fgithub.com\u002Fimran-saleemi)\u003C\u002Fli> \u003Cli>[Soumyadip Sengupta](https:\u002F\u002Fgithub.com\u002Fsenguptaumd)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a1152adfc4dc.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FWACV51458.2022.00319) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPeterL1n\u002FRobustVideoMatting?style=social)](https:\u002F\u002Fgithub.com\u002FPeterL1n\u002FRobustVideoMatting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.11515)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpeterl1n.github.io\u002FRobustVideoMatting)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Fpdbpmg\u002Fr_robust_highresolution_video_matting_with\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJvzltozpbpk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAy-mGCEYEzM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVL-0K6HjhvQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJhuf6M_VrBI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_oN9yyRi3HY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F10z-pNKRnVNsp0Lq9tH1J_XPZ7CBC_uHm) | 24.11.2021 |\n| AnimeGANv2 | An improved version of AnimeGAN - it prevents the generation of high-frequency artifacts by simply changing the normalization of features in the network | \u003Cul>\u003Cli>[Xin Chen](https:\u002F\u002Fgithub.com\u002FTachibanaYoshino)\u003C\u002Fli> \u003Cli>[Gang Liu](https:\u002F\u002Fgithub.com\u002Flg0061408)\u003C\u002Fli> \u003Cli>[bryandlee](https:\u002F\u002Fgithub.com\u002Fbryandlee)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c49bddb96517.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-981-15-5577-0_18) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fbryandlee\u002Fanimegan2-pytorch?style=social)](https:\u002F\u002Fgithub.com\u002Fbryandlee\u002Fanimegan2-pytorch) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTachibanaYoshino\u002FAnimeGANv2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTachibanaYoshino\u002FAnimeGAN)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002FAnimeGANv2)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftachibanayoshino.github.io\u002FAnimeGANv2\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fbryandlee\u002Fanimegan2-pytorch\u002Fblob\u002Fmaster\u002Fcolab_demo.ipynb) | 17.11.2021 |\n| SOAT | StyleGAN of All Trades: Image Manipulation with Only Pretrained StyleGAN | \u003Cul>\u003Cli>[Min Jin Chong](https:\u002F\u002Fmchong6.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hsin-Ying Lee](http:\u002F\u002Fhsinyinglee.com\u002F)\u003C\u002Fli> \u003Cli>[David Forsyth](http:\u002F\u002Fluthuli.cs.uiuc.edu\u002F~daf\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmchong6\u002FSOAT?style=social)](https:\u002F\u002Fgithub.com\u002Fmchong6\u002FSOAT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.01619)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjustinpinkney\u002Ftoonify), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002FSOAT)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmchong6\u002FSOAT\u002Fblob\u002Fmaster\u002Finfinity.ipynb) | 13.11.2021 |\n| Arnheim | Generative Art Using Neural Visual Grammars and Dual Encoders | \u003Cul>\u003Cli>[Chrisantha Fernando](https:\u002F\u002Fwww.chrisantha.co.uk\u002F)\u003C\u002Fli> \u003Cli>[Ali Eslami](http:\u002F\u002Farkitus.com\u002F)\u003C\u002Fli> \u003Cli>[Jean-Baptiste Alayrac](https:\u002F\u002Fwww.jbalayrac.com\u002F)\u003C\u002Fli> \u003Cli>[Piotr Mirowski](https:\u002F\u002Fpiotrmirowski.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dylan Banarse](https:\u002F\u002Fwww.2ne1.com\u002F)\u003C\u002Fli> \u003Cli>[Simon Osindero](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Jq8ZS5kAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepmind\u002Farnheim?style=social)](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Farnheim) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.00162), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.14843), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.07729), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1606.02580), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1609.09106)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fdall-e)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FCompositional_pattern-producing_network)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=U7guaMdeF4g), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zh0goLbS-l0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=SYJGNt7yu6M), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=MxkYKa0x5AU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Farnheim\u002Fblob\u002Fmaster\u002Farnheim_2.ipynb) | 11.11.2021 |\n| StyleGAN 2 | Generation of faces, cars, etc. | [Mikael Christensen](https:\u002F\u002Fgithub.com\u002FSyntopia) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ada1741909ed.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00813) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNVlabs\u002Fstylegan2?style=social)](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F1912.04958)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fc-NJtV9Jvp0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1ShgW6wohEFQtqs_znMna3dzrcVoABKIH) | 05.11.2021 |\n| ByteTrack | Multi-Object Tracking by Associating Every Detection Box | \u003Cul>\u003Cli>[Yifu Zhang](https:\u002F\u002Fgithub.com\u002Fifzhang)\u003C\u002Fli> \u003Cli>[Peize Sun](https:\u002F\u002Fpeizesun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yi Jiang](https:\u002F\u002Fgithub.com\u002FiFighting)\u003C\u002Fli> \u003Cli>[Dongdong Yu](https:\u002F\u002Fmiracle-fmh.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ping Luo](http:\u002F\u002Fluoping.me\u002F)\u003C\u002Fli> \u003Cli>[Xinggang Wang](https:\u002F\u002Fxinggangw.info\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cf78940e823f.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20047-2_1) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fifzhang\u002FByteTrack?style=social)](https:\u002F\u002Fgithub.com\u002Fifzhang\u002FByteTrack) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2110.06864)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fmotchallenge.net\u002F), [data](https:\u002F\u002Fwww.crowdhuman.org\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMegvii-BaseDetection\u002FYOLOX), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fifzhang\u002FFairMOT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FPeizeSun\u002FTransTrack), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsamylee\u002FTowards-Realtime-MOT-Cpp)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Ftask\u002Fmulti-object-tracking)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1bDilg4cmXFa8HCKHbsZ_p16p0vrhLyu0) | 30.10.2021 |\n| GPT-2 | Retrain an advanced text generating neural network on any text dataset using gpt-2-simple! | [Max Woolf](https:\u002F\u002Fminimaxir.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fopenai\u002Fgpt-2?style=social)](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fgpt-2) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fminimaxir.com\u002F2019\u002F09\u002Fhowto-gpt2\u002F), [blog post](https:\u002F\u002Fopenai.com\u002Fresearch\u002Fbetter-language-models)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fminimaxir\u002Fgpt-2-simple)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Faqlzde\u002Fr_openai_better_language_models_and_their\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1VLG8e7YSEwypxU-noRNhsv5dW4NfTGce) | 18.10.2021 |\n| ConvMixer | An extremely simple model that is similar in spirit to the ViT and the even-more-basic MLP-Mixer in that it operates directly on patches as input, separates the mixing of spatial and channel dimensions, and maintains equal size and resolution throughout the network | \u003Cul>\u003Cli>[Asher Trockman](http:\u002F\u002Fashertrockman.com\u002F)\u003C\u002Fli> \u003Cli>[Zico Kolter](http:\u002F\u002Fzicokolter.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Flocuslab\u002Fconvmixer?style=social)](https:\u002F\u002Fgithub.com\u002Flocuslab\u002Fconvmixer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.09792)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flocuslab\u002Fconvmixer-cifar10), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frwightman\u002Fpytorch-image-models)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fcodex\u002Fan-overview-on-convmixer-patches-are-all-you-need-8502a8d87011)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGl0s0GDqN3c?t=990)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Flocuslab\u002Fconvmixer\u002Fblob\u002Fmain\u002Fpytorch-image-models\u002Fnotebooks\u002FEffResNetComparison.ipynb) | 06.10.2021 |\n| IC-GAN | Instance-Conditioned GAN | \u003Cul>\u003Cli>[Arantxa Casanova](https:\u002F\u002Fgithub.com\u002FArantxaCasanova)\u003C\u002Fli> \u003Cli>[Marlène Careil](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fmarl%C3%A8ne-careil-901804155)\u003C\u002Fli> \u003Cli>[Jakob Verbeek](http:\u002F\u002Fthoth.inrialpes.fr\u002F~verbeek\u002F)\u003C\u002Fli> \u003Cli>[Michał Drożdżal](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=XK_ktwQAAAAJ)\u003C\u002Fli> \u003Cli>[Adriana Romero-Soriano](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fadriromsor)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fic_gan?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fic_gan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.05070)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fai.facebook.com\u002Fblog\u002Finstance-conditioned-gans\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffaiss), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fajbrock\u002FBigGAN-PyTorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbioinf-jku\u002FTTUR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fdata-efficient-gans)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2021\u002Fhash\u002Fe7ac288b0f2d41445904d071ba37aaff-Abstract.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fic_gan\u002Fblob\u002Fmaster\u002Finference\u002Ficgan_colab.ipynb) | 01.10.2021 |\n| Skillful Precipitation Nowcasting Using Deep Generative Models of Radar | Open-sourced dataset and model snapshot for precipitation nowcasting | \u003Cul>\u003Cli>[Suman Ravuri](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsuman-ravuri-81928082)\u003C\u002Fli> \u003Cli>[Karel Lenc](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~karel\u002F)\u003C\u002Fli> \u003Cli>[Matthew Willson](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fmatthew-willson-6a1b422)\u003C\u002Fli> \u003Cli>[Dmitry Kangin](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vv-leaMAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Rémi Lam](https:\u002F\u002Fgithub.com\u002Fremilam)\u003C\u002Fli> \u003Cli>[Piotr Mirowski](https:\u002F\u002Fpiotrmirowski.com\u002F)\u003C\u002Fli> \u003Cli>[Maria Athanassiadou](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=VtkgHP0AAAAJ)\u003C\u002Fli> \u003Cli>[Sheleem Kashem](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsheleemkashem\u002F)\u003C\u002Fli> \u003Cli>[Rachel Prudden](https:\u002F\u002Fcomputerscience.exeter.ac.uk\u002Fstaff\u002Frep218)\u003C\u002Fli> \u003Cli>[Amol Mandhane](https:\u002F\u002Fgithub.com\u002Famol-mandhane)\u003C\u002Fli> \u003Cli>[Aidan Clark](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=_19DrfIAAAAJ)\u003C\u002Fli> \u003Cli>[Andrew Brock](https:\u002F\u002Fgithub.com\u002Fajbrock)\u003C\u002Fli> \u003Cli>[Karen Simonyan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=L7lMQkQAAAAJ)\u003C\u002Fli> \u003Cli>[Raia Hadsell](https:\u002F\u002Fgithub.com\u002Fraiah)\u003C\u002Fli> \u003Cli>[Niall Robinson](https:\u002F\u002Fgithub.com\u002Fniallrobinson)\u003C\u002Fli> \u003Cli>[Ellen Clancy](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fellen-clancy-815967124)\u003C\u002Fli> \u003Cli>[Shakir Mohamed](https:\u002F\u002Fwww.shakirm.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_5fb907f53a5e.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41586-021-03854-z) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepmind\u002Fdeepmind-research?style=social)](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fdeepmind-research\u002Ftree\u002Fmaster\u002Fnowcasting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.00954)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fdeepmind.com\u002Fblog\u002Farticle\u002Fnowcasting)\u003C\u002Fli>\u003Cli>[local kernel](https:\u002F\u002Fresearch.google.com\u002Fcolaboratory\u002Flocal-runtimes.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fhub)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Fdeepmind-research\u002Fblob\u002Fmaster\u002Fnowcasting\u002FOpen_sourced_dataset_and_model_snapshot_for_precipitation_nowcasting.ipynb) | 29.09.2021 |\n| Live Speech Portraits | Real-Time Photorealistic Talking-Head Animation | \u003Cul>\u003Cli>[Yuanxun Lu](https:\u002F\u002Fgithub.com\u002FYuanxunLu)\u003C\u002Fli> \u003Cli>[Jinxiang Chai](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=OcN1_gwAAAAJ)\u003C\u002Fli> \u003Cli>[Xun Cao](https:\u002F\u002Fcite.nju.edu.cn\u002FPeople\u002FFaculty\u002F20190621\u002Fi5054.html)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c42b73ecb91d.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3478513.3480484) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYuanxunLu\u002FLiveSpeechPortraits?style=social)](https:\u002F\u002Fgithub.com\u002FYuanxunLu\u002FLiveSpeechPortraits) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.10595)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flelechen63\u002FATVGnet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flelechen63\u002FTalking-head-Generation-with-Rhythmic-Head-Motion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDinoMan\u002Fspeech-driven-animation), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjunyanz\u002Fpytorch-CycleGAN-and-pix2pix)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyuanxunlu.github.io\u002Fprojects\u002FLiveSpeechPortraits\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1tKvi-9kY3GkEK8lgtfTSM70rMFo_TY50) | 26.09.2021 |\n| StylEx | Training a GAN to explain a classifier in StyleSpace | \u003Cul>\u003Cli>[Oran Lang](https:\u002F\u002Fresearch.google\u002Fpeople\u002F105975\u002F)\u003C\u002Fli> \u003Cli>[Yossi Gandelsman](https:\u002F\u002Fyossigandelsman.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michal Yarom](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=GMVxiYgAAAAJ)\u003C\u002Fli> \u003Cli>[Yoav Wald](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=hh5nOn4AAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Gal Elidan](https:\u002F\u002Fresearch.google\u002Fpeople\u002F105719\u002F)\u003C\u002Fli> \u003Cli>[Avinatan Hassidim](https:\u002F\u002Fresearch.google\u002Fpeople\u002F105831\u002F)\u003C\u002Fli> \u003Cli>[William Freeman](https:\u002F\u002Fbillf.mit.edu\u002F)\u003C\u002Fli> \u003Cli>[Phillip Isola](http:\u002F\u002Fweb.mit.edu\u002Fphillipi\u002F)\u003C\u002Fli> \u003Cli>[Amir Globerso](https:\u002F\u002Fcs3801.wixsite.com\u002Famirgloberson)\u003C\u002Fli> \u003Cli>[Michal Irani](http:\u002F\u002Fwww.weizmann.ac.il\u002Fmath\u002Firani\u002F)\u003C\u002Fli> \u003Cli>[Inbar Mosseri](https:\u002F\u002Fresearch.google\u002Fpeople\u002FInbarMosseri\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_50f212994a1a.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00073) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fexplaining-in-style?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fexplaining-in-style) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.13369), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.10112), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.12799), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1912.04958), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1710.01711)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fai.googleblog.com\u002F2022\u002F01\u002Fintroducing-stylex-new-approach-for.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fexplaining-in-style.github.io\u002F)\u003C\u002Fli>\u003Cli>[supplementary](https:\u002F\u002Fexplaining-in-style.github.io\u002Fsupmat.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FwLk2eBdXH4M)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fexplaining-in-style\u002Fblob\u002Fmain\u002FExplaining_in_Style_AttFind.ipynb) | 25.08.2021 |\n| VITS | Parallel end-to-end TTS method that generates more natural sounding audio than current two-stage models | \u003Cul>\u003Cli>[Jaehyeon Kim](https:\u002F\u002Fjaywalnut310.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jungil Kong](https:\u002F\u002Fgithub.com\u002Fjik876)\u003C\u002Fli> \u003Cli>[Juhee Son](https:\u002F\u002Fjuheeuu.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjaywalnut310\u002Fvits?style=social)](https:\u002F\u002Fgithub.com\u002Fjaywalnut310\u002Fvits) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.06103)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fjaywalnut310.github.io\u002Fvits-demo\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1CO61pZizDj7en71NQG_aqqKdGaA_SaBf) | 23.08.2021 |\n| Bringing Old Photo Back to Life | Restoring old photos that suffer from severe degradation through a deep learning approach | \u003Cul>\u003Cli>[Ziyu Wan](http:\u002F\u002Fraywzy.com\u002F)\u003C\u002Fli> \u003Cli>[Bo Zhang](https:\u002F\u002Fbo-zhang.me\u002F)\u003C\u002Fli> \u003Cli>[Dongdong Chen](http:\u002F\u002Fwww.dongdongchen.bid\u002F)\u003C\u002Fli> \u003Cli>[Pan Zhang](https:\u002F\u002Fpanzhang0212.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dong Chen](http:\u002F\u002Fwww.dongchen.pro\u002F)\u003C\u002Fli> \u003Cli>[Jing Liao](https:\u002F\u002Fliaojing.github.io\u002Fhtml\u002F)\u003C\u002Fli> \u003Cli>[Fang Wen](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Ffangwen\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_5cd53ffe91f1.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00282) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002FBringing-Old-Photos-Back-to-Life?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FBringing-Old-Photos-Back-to-Life) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.09484)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Freplicate.com\u002Fmicrosoft\u002Fbringing-old-photos-back-to-life)\u003C\u002Fli>\u003Cli>[project](http:\u002F\u002Fraywzy.com\u002FOld_Photo\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQ5bhszQq9eA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1NEm6AsybIiC5TwTU_4DqDkQO0nFRB-uA) | 13.07.2021 |\n| PTI | Pivotal Tuning Inversion enables employing off-the-shelf latent based semantic editing techniques on real images using StyleGAN | \u003Cul>\u003Cli>[Daniel Roich](https:\u002F\u002Fgithub.com\u002Fdanielroich)\u003C\u002Fli> \u003Cli>[Ron Mokady](https:\u002F\u002Frmokady.github.io\u002F)\u003C\u002Fli> \u003Cli>[Amit Bermano](https:\u002F\u002Fwww.cs.tau.ac.il\u002F~amberman\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_16de55b38de6.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3544777) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdanielroich\u002FPTI?style=social)](https:\u002F\u002Fgithub.com\u002Fdanielroich\u002FPTI) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.05744)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frichzhang\u002FPerceptualSimilarity)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdanielroich\u002FPTI\u002Fblob\u002Fmain\u002Fnotebooks\u002Finference_playground.ipynb) | 01.07.2021 |\n| TediGAN | Framework for multi-modal image generation and manipulation with textual descriptions | \u003Cul>\u003Cli>[Weihao Xia](https:\u002F\u002Fgithub.com\u002Fweihaox)\u003C\u002Fli> \u003Cli>[Yujiu Yang](http:\u002F\u002Fwww.fiesta.tsinghua.edu.cn\u002Fpi\u002F3\u002F24)\u003C\u002Fli> \u003Cli>[Jing-Hao Xue](http:\u002F\u002Fwww.homepages.ucl.ac.uk\u002F~ucakjxu\u002F)\u003C\u002Fli> \u003Cli>[Baoyuan Wu](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fbaoyuanwu2015\u002Fhome)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_dbc9ae17c37e.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00229) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FIIGROUP\u002FTediGAN?style=social)](https:\u002F\u002Fgithub.com\u002FIIGROUP\u002FTediGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.03308), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.08910)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fweihaox\u002FMulti-Modal-CelebA-HQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch\u002F), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffyu\u002Flsun)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FL8Na2f5viAM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](http:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fweihaox\u002FTediGAN\u002Fblob\u002Fmaster\u002Fplayground.ipynb) | 30.06.2021 |\n| SCALE | Modeling Clothed Humans with a Surface Codec of Articulated Local Elements | \u003Cul>\u003Cli>[Qianli Ma](https:\u002F\u002Fqianlim.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shunsuke Saito](https:\u002F\u002Fshunsukesaito.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jinlong Yang](https:\u002F\u002Fis.mpg.de\u002F~jyang)\u003C\u002Fli> \u003Cli>[Siyu Tang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=BUDh_4wAAAAJ)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_380e811aec56.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01582) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fqianlim\u002FSCALE?style=social)](https:\u002F\u002Fgithub.com\u002Fqianlim\u002FSCALE) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.07660)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcape.is.tue.mpg.de\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkrrish94\u002Fchamferdist), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fshunsukesaito\u002FSCANimate)\u003C\u002Fli>\u003Cli>[poster](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F650\u002FSCALE_poster_CVPR_final_compressed.pdf)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fqianlim.github.io\u002FSCALE.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-EvWqFCUb7U), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fv4rWCxJJzhc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1lp6r-A-s1kBorIvg6rLD4Ja3o6JOvu3G) | 26.06.2021 |\n| CogView | Mastering Text-to-Image Generation via Transformers | \u003Cul>\u003Cli>[Ming Ding](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Va50YzkAAAAJ)\u003C\u002Fli> \u003Cli>[Zhuoyi Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=tgAt-gEAAAAJ)\u003C\u002Fli> \u003Cli>[Wenyi Hong](https:\u002F\u002Fgithub.com\u002Fwenyihong)\u003C\u002Fli> \u003Cli>[Wendi Zheng](https:\u002F\u002Fgithub.com\u002Fminkowski0125)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chang Zhou](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QeSoG3sAAAAJ)\u003C\u002Fli> \u003Cli>[Junyang Lin](https:\u002F\u002Fjustinlin610.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xu Zou](http:\u002F\u002Fxuzou.cn\u002F)\u003C\u002Fli> \u003Cli>[Zhou Shao](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FShao_Zhou4)\u003C\u002Fli> \u003Cli>[Hongxia Yang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fhystatistics\u002Fhome)\u003C\u002Fli> \u003Cli>[Jie Tang](https:\u002F\u002Fkeg.cs.tsinghua.edu.cn\u002Fjietang\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTHUDM\u002FCogView?style=social)](https:\u002F\u002Fgithub.com\u002FTHUDM\u002FCogView) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.13290)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fthudm.github.io\u002FCogView\u002Findex.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fapex), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FSleepychord\u002Fcogdata)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Ftowardsdatascience.com\u002Fcogview-image-generation-and-language-modelling-at-scale-8d358a0686d2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2021\u002Fhash\u002Fa4d92e2cd541fca87e4620aba658316d-Abstract.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Fnmxsd8\u002Fr_cogview_mastering_texttoimage_generation_via\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FCw1r8ACIj8U)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Bi2TnSUp2vNiSUhamsNuC4HqkZ2J4WwZ) | 21.06.2021 |\n| GANs N' Roses | Stable, Controllable, Diverse Image to Image Translation | \u003Cul>\u003Cli>[Min Jin Chong](https:\u002F\u002Fmchong6.github.io\u002F)\u003C\u002Fli> \u003Cli>[David Forsyth](http:\u002F\u002Fluthuli.cs.uiuc.edu\u002F~daf\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmchong6\u002FGANsNRoses?style=social)](https:\u002F\u002Fgithub.com\u002Fmchong6\u002FGANsNRoses) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.06561), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2007.06600)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fznxlwm\u002FUGATIT-pytorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVNg0NyCGl_4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmchong6\u002FGANsNRoses\u002Fblob\u002Fmaster\u002Finference_colab.ipynb) | 19.06.2021 |\n| Rethinking Style Transfer: From Pixels to Parameterized Brushstrokes | A method to stylize images by optimizing parameterized brushstrokes instead of pixels | \u003Cul>\u003Cli>[Dmytro Kotovenko](https:\u002F\u002Fscholar.google.de\u002Fcitations?user=T_U8yxwAAAAJ)\u003C\u002Fli> \u003Cli>[Matthias Wright](https:\u002F\u002Fmatthias-wright.github.io\u002F)\u003C\u002Fli> \u003Cli>[Arthur Heimbrecht](https:\u002F\u002Fgithub.com\u002Farwehei)\u003C\u002Fli> \u003Cli>[Björn Ommer](https:\u002F\u002Fommer-lab.com\u002Fpeople\u002Fommer\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_61bdf3918c27.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01202) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCompVis\u002Fbrushstroke-parameterized-style-transfer?style=social)](https:\u002F\u002Fgithub.com\u002FCompVis\u002Fbrushstroke-parameterized-style-transfer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.17185)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fcompvis.github.io\u002Fbrushstroke-parameterized-style-transfer\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FCompVis\u002Fbrushstroke-parameterized-style-transfer\u002Fblob\u002Ftensorflow_v2\u002Fnotebooks\u002FBrushstrokeStyleTransfer_TF2.ipynb) | 02.06.2021 |\n| Pixel2Style2Pixel | Encoding in Style: A StyleGAN Encoder for Image-to-Image Translation | \u003Cul>\u003Cli>[Elad Richardson](https:\u002F\u002Fgithub.com\u002Feladrich)\u003C\u002Fli> \u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yotam Nitzan](https:\u002F\u002Fyotamnitzan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_4a5992da112b.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00232) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Feladrich\u002Fpixel2style2pixel?style=social)](https:\u002F\u002Fgithub.com\u002Feladrich\u002Fpixel2style2pixel) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.00951)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHuangYG123\u002FCurricularFace)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Feladrich.github.io\u002Fpixel2style2pixel\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbfvSwhqsTgM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Feladrich\u002Fpixel2style2pixel\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 01.06.2021 |\n| Fine-tuning a BERT | We will work through fine-tuning a BERT model using the tensorflow-models PIP package | \u003Cul>\u003Cli>[Chen Chen](https:\u002F\u002Fgithub.com\u002FchenGitHuber)\u003C\u002Fli> \u003Cli>[Claire Yao](https:\u002F\u002Fgithub.com\u002Fclaireyao-fen)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1d8bb9ef5451.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002FN19-1423) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.04805)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Ftensorflow.org\u002Fhub)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fmodels\u002Fblob\u002Fmaster\u002Fofficial\u002Fcolab\u002Ffine_tuning_bert.ipynb) | 25.05.2021 |\n| ReStyle | A Residual-Based StyleGAN Encoder via Iterative Refinement | \u003Cul>\u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_4b939d0bae6b.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00664) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyuval-alaluf\u002Frestyle-encoder?style=social)](https:\u002F\u002Fgithub.com\u002Fyuval-alaluf\u002Frestyle-encoder) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.02699), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.00951), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.02766)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTreB1eN\u002FInsightFace_Pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyuval-alaluf.github.io\u002Frestyle-encoder\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyuval-alaluf\u002Frestyle-encoder\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 21.05.2021 |\n| Motion Representations for Articulated Animation | Novel motion representations for animating articulated objects consisting of distinct parts | \u003Cul>\u003Cli>[Aliaksandr Siarohin](https:\u002F\u002Faliaksandrsiarohin.github.io\u002Faliaksandr-siarohin-website\u002F)\u003C\u002Fli> \u003Cli>[Oliver Woodford](https:\u002F\u002Fojwoodford.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jian Ren](https:\u002F\u002Falanspike.github.io\u002F)\u003C\u002Fli> \u003Cli>[Menglei Chai](https:\u002F\u002Fmlchai.com\u002F)\u003C\u002Fli> \u003Cli>[Sergey Tulyakov](http:\u002F\u002Fwww.stulyakov.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_852569f5d4b8.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01344) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsnap-research\u002Farticulated-animation?style=social)](https:\u002F\u002Fgithub.com\u002Fsnap-research\u002Farticulated-animation) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.11280)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsnap-research.github.io\u002Farticulated-animation\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=gpBYN8t8_yY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAliaksandrSiarohin\u002Farticulated-animation\u002Fblob\u002Fmaster\u002Fdemo.ipynb) | 29.04.2021 |\n| SAM | Age Transformation Using a Style-Based Regression Model | \u003Cul>\u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0239aadbb33a.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3450626.3459805) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyuval-alaluf\u002FSAM?style=social)](https:\u002F\u002Fgithub.com\u002Fyuval-alaluf\u002FSAM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.02754)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Feladrich\u002Fpixel2style2pixel), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyuval-alaluf.github.io\u002FSAM\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FX_pYC_LtBFw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](http:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyuval-alaluf\u002FSAM\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fanimation_inference_playground.ipynb) | 26.04.2021 |\n| Geometry-Free View Synthesis | Is a geometric model required to synthesize novel views from a single image? | \u003Cul>\u003Cli>[Robin Rombach](https:\u002F\u002Fgithub.com\u002Frromb)\u003C\u002Fli> \u003Cli>[Patrick Esser](https:\u002F\u002Fgithub.com\u002Fpesser)\u003C\u002Fli> \u003Cli>[Björn Ommer](https:\u002F\u002Fommer-lab.com\u002Fpeople\u002Fommer\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9f50cfdb8dc3.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.01409) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCompVis\u002Fgeometry-free-view-synthesis?style=social)](https:\u002F\u002Fgithub.com\u002FCompVis\u002Fgeometry-free-view-synthesis) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.07652)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fgoogle.github.io\u002Frealestate10k\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcolmap\u002Fcolmap)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FCompVis\u002Fgeometry-free-view-synthesis\u002Fblob\u002Fmaster\u002Fscripts\u002Fbraindance.ipynb) | 22.04.2021 |\n| NeRViS | An algorithm for full-frame video stabilization by first estimating dense warp fields | \u003Cul>\u003Cli>[Yu-Lun Liu](http:\u002F\u002Fwww.cmlab.csie.ntu.edu.tw\u002F~yulunliu\u002F)\u003C\u002Fli> \u003Cli>[Wei-Sheng Lai](https:\u002F\u002Fwww.wslai.net\u002F)\u003C\u002Fli> \u003Cli>[Ming-Hsuan Yang](https:\u002F\u002Ffaculty.ucmerced.edu\u002Fmhyang\u002F)\u003C\u002Fli> \u003Cli>[Yung-Yu Chuang](https:\u002F\u002Fwww.csie.ntu.edu.tw\u002F~cyy\u002F)\u003C\u002Fli> \u003Cli>[Jia-Bin Huang](https:\u002F\u002Fjbhuang0604.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fd196cdcdbc9.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00230) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Falex04072000\u002FNeRViS?style=social)](https:\u002F\u002Fgithub.com\u002Falex04072000\u002FNeRViS) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.06205)\u003C\u002Fli>\u003Cli>[data](http:\u002F\u002Fliushuaicheng.org\u002FSIGGRAPH2013\u002Fdatabase.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcxjyxxme\u002Fdeep-online-video-stabilization), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjinsc37\u002FDIFRINT)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Falex04072000.github.io\u002FNeRViS\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FKO3sULs4hso)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1l-fUzyM38KJMZyKMBWw_vu7ZUyDwgdYH) | 11.04.2021 |\n| NeX | View synthesis based on enhancements of multiplane image that can reproduce NeXt-level view-dependent effects in real time | \u003Cul>\u003Cli>[Suttisak Wizadwongsa](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsuttisak-wizadwongsa-763a931a5\u002F)\u003C\u002Fli> \u003Cli>[Pakkapon Phongthawee](http:\u002F\u002Fpureexe.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiraphon Yenphraphai](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fjiraphon-yenphraphai-990ba6175\u002F)\u003C\u002Fli> \u003Cli>[Supasorn Suwajanakorn](https:\u002F\u002Fwww.supasorn.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c42449f21921.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00843) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fnex-mpi\u002Fnex-code?style=social)](https:\u002F\u002Fgithub.com\u002Fnex-mpi\u002Fnex-code) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.05606)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fvistec-my.sharepoint.com\u002Fpersonal\u002Fpakkapon_p_s19_vistec_ac_th\u002F_layouts\u002F15\u002Fonedrive.aspx?id=%2Fpersonal%2Fpakkapon%5Fp%5Fs19%5Fvistec%5Fac%5Fth%2FDocuments%2Fpublic%2FVLL%2FNeX%2Fshiny%5Fdatasets&originalPath=aHR0cHM6Ly92aXN0ZWMtbXkuc2hhcmVwb2ludC5jb20vOmY6L2cvcGVyc29uYWwvcGFra2Fwb25fcF9zMTlfdmlzdGVjX2FjX3RoL0VuSVVoc1JWSk9kTnNaXzRzbWRoeWUwQjh6MFZseHFPUjM1SVIzYnAwdUd1cFE%5FcnRpbWU9WXRVQTQtQTcyVWc), [data](https:\u002F\u002Fvistec-my.sharepoint.com\u002Fpersonal\u002Fpakkapon_p_s19_vistec_ac_th\u002F_layouts\u002F15\u002Fonedrive.aspx?originalPath=aHR0cHM6Ly92aXN0ZWMtbXkuc2hhcmVwb2ludC5jb20vOmY6L2cvcGVyc29uYWwvcGFra2Fwb25fcF9zMTlfdmlzdGVjX2FjX3RoL0VyalBSUkw5Sm5GSXA4TU42ZDFqRXVvQjNYVm94SmtmZlBqZm9QeWhIa2owZGc%5FcnRpbWU9bC0yYWctRTcyVWc&id=%2Fpersonal%2Fpakkapon%5Fp%5Fs19%5Fvistec%5Fac%5Fth%2FDocuments%2Fpublic%2FVLL%2FNeX%2Fmodified%5Fdataset)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFyusion\u002FLLFF)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnex-mpi.github.io\u002F)\u003C\u002Fli>\u003Cli>[vistec](https:\u002F\u002Fvistec.ist\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=HyfkF7Z-ddA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1hXVvYdAwLA0EFg2zrafJUE0bFgB_F7PU) | 25.03.2021 |\n| Score SDE | Score-Based Generative Modeling through Stochastic Differential Equations | \u003Cul>\u003Cli>[Yang Song](https:\u002F\u002Fyang-song.net\u002F)\u003C\u002Fli> \u003Cli>[Jascha Sohl-Dickstein](http:\u002F\u002Fwww.sohldickstein.com\u002F)\u003C\u002Fli> \u003Cli>[Diederik Kingma](http:\u002F\u002Fdpkingma.com\u002F)\u003C\u002Fli> \u003Cli>[Abhishek Kumar](https:\u002F\u002Fabhishek.umiacs.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Stefano Ermon](https:\u002F\u002Fcs.stanford.edu\u002F~ermon\u002F)\u003C\u002Fli> \u003Cli>[Ben Poole](https:\u002F\u002Fcs.stanford.edu\u002F~poole\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyang-song\u002Fscore_sde?style=social)](https:\u002F\u002Fgithub.com\u002Fyang-song\u002Fscore_sde) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.13456), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.05600), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.09011), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.11239)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyang-song\u002Fscore_sde_pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fml_collections)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FL9ZegT87QK8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyang-song\u002Fscore_sde\u002Fblob\u002Fmain\u002FScore_SDE_demo.ipynb) | 18.03.2021 |\n| Talking Head Anime from a Single Image | The network takes as input an image of an anime character's face and a desired pose, and it outputs another image of the same character in the given pose | [Pramook Khungurn](https:\u002F\u002Fpkhungurn.github.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fpkhungurn\u002Ftalking-head-anime-demo?style=social)](https:\u002F\u002Fgithub.com\u002Fpkhungurn\u002Ftalking-head-anime-demo) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flincolnhard\u002Fhead-pose-estimation)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpkhungurn.github.io\u002Ftalking-head-anime\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FVirtual_YouTuber), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FMikuMikuDance)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FkMQCERkTdO0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FT1Gp-RxFZwU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FFioRJ6x_RbI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fpkhungurn\u002Ftalking-head-anime-demo\u002Fblob\u002Fmaster\u002Ftha_colab.ipynb) | 23.02.2021 |\n| NFNet | An adaptive gradient clipping technique, a significantly improved class of Normalizer-Free ResNets | \u003Cul>\u003Cli>[Andrew Brock](https:\u002F\u002Fgithub.com\u002Fajbrock)\u003C\u002Fli> \u003Cli>[Soham De](https:\u002F\u002Fsohamde.github.io\u002F)\u003C\u002Fli> \u003Cli>[Samuel L. Smith](https:\u002F\u002Fscholar.google.co.uk\u002Fcitations?user=fyEqU5oAAAAJ)\u003C\u002Fli> \u003Cli>[Karen Simonyan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=L7lMQkQAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepmind\u002Fdeepmind-research?style=social)](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fdeepmind-research\u002Ftree\u002Fmaster\u002Fnfnets) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.06171), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2101.08692)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fjaxline)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FrNkHjZtH0RQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002Fqyy2WhRRSI4?feature=share)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Fdeepmind-research\u002Fblob\u002Fmaster\u002Fnfnets\u002Fnfnet_demo_colab.ipynb) | 17.02.2021 |\n| RITM | Simple feedforward model for click-based interactive segmentation that employs the segmentation masks from previous steps | \u003Cul>\u003Cli>[Konstantin Sofiiuk](https:\u002F\u002Fgithub.com\u002Fksofiyuk)\u003C\u002Fli> \u003Cli>[Ilia Petrov](https:\u002F\u002Fvirtualhumans.mpi-inf.mpg.de\u002Fpeople\u002FPetrov.html)\u003C\u002Fli> \u003Cli>[Anton Konushin](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ZT_k-wMAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_81755896538e.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICIP46576.2022.9897365) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsupervisely-ecosystem\u002Fritm-interactive-segmentation?style=social)](https:\u002F\u002Fgithub.com\u002Fsupervisely-ecosystem\u002Fritm-interactive-segmentation) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.06583)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHRNet\u002FHRNet-Image-Classification)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Finteractive-segmentation-on-grabcut?p=reviving-iterative-training-with-mask), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Finteractive-segmentation-on-berkeley?p=reviving-iterative-training-with-mask)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsupervisely-ecosystem\u002Fritm_interactive_segmentation\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fcolab_test_any_model.ipynb) | 13.02.2021 |\n| CLIP | A neural network which efficiently learns visual concepts from natural language supervision | \u003Cul>\u003Cli>[Jong Wook Kim](https:\u002F\u002Fjongwook.kim\u002F)\u003C\u002Fli> \u003Cli>[Alec Radford](http:\u002F\u002Fnewmu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ilya Sutskever](http:\u002F\u002Fwww.cs.utoronto.ca\u002F~ilya\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fopenai\u002FCLIP?style=social)](https:\u002F\u002Fgithub.com\u002Fopenai\u002FCLIP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.00020)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fwww.cs.toronto.edu\u002F~kriz\u002Fcifar.html)\u003C\u002Fli>\u003Cli>[paper](https:\u002F\u002Fcdn.openai.com\u002Fpapers\u002FLearning_Transferable_Visual_Models_From_Natural_Language_Supervision.pdf)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fopenai.com\u002Fblog\u002Fclip\u002F)\u003C\u002Fli>\u003Cli>[slides](https:\u002F\u002Ficml.cc\u002Fmedia\u002Ficml-2021\u002FSlides\u002F9193.pdf)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fopenai\u002Fclip\u002Fblob\u002Fmaster\u002FInteracting_with_CLIP.ipynb) | 29.01.2021 |\n| Adversarial Patch | A method to create universal, robust, targeted adversarial image patches in the real world | [Tom Brown](https:\u002F\u002Fgithub.com\u002Fnottombrown) | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1712.09665)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcleverhans-lab\u002Fcleverhans\u002Fblob\u002Fmaster\u002Fexamples\u002Fadversarial_patch\u002FAdversarialPatch.ipynb) | 27.01.2021 |\n| MSG-Net | Multi-style Generative Network with a novel Inspiration Layer, which retains the functionality of optimization-based approaches and has the fast speed of feed-forward networks | \u003Cul>\u003Cli>[Hang Zhang](https:\u002F\u002Fhangzhang.org\u002F)\u003C\u002Fli> \u003Cli>[Kristin Dana](https:\u002F\u002Fwww.ece.rutgers.edu\u002F~kdana\u002Fdana.html)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_81755896538e.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-11018-5_32) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1703.06953)\u003C\u002Fli>\u003Cli>[project](http:\u002F\u002Fcomputervisionrutgers.github.io\u002FMSG-Net\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=oy6pWNWBt4Y)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fzhanghang1989\u002FPyTorch-Multi-Style-Transfer\u002Fblob\u002Fmaster\u002Fmsgnet.ipynb) | 25.01.2021 |\n| Neural Style Transfer | Implementation of Neural Style Transfer in Keras 2.0+ | [Somshubra Majumdar](http:\u002F\u002Ftitu1994.github.io\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_73994007fd6b.png)](https:\u002F\u002Fdoi.org\u002F10.1167\u002F16.12.326) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftitu1994\u002FNeural-Style-Transfer?style=social)](https:\u002F\u002Fgithub.com\u002Ftitu1994\u002FNeural-Style-Transfer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F1508.06576), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F1605.04603), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1606.05897)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftitu1994\u002FNeural-Style-Transfer\u002Fblob\u002Fmaster\u002FNeuralStyleTransfer.ipynb) | 22.01.2021 |\n| SkyAR | A vision-based method for video sky replacement and harmonization, which can automatically generate realistic and dramatic sky backgrounds in videos with controllable styles | [Zhengxia Zou](http:\u002F\u002Fwww-personal.umich.edu\u002F~zzhengxi\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_612548bd9a37.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTIP.2022.3192717) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjiupinjia\u002FSkyAR?style=social)](https:\u002F\u002Fgithub.com\u002Fjiupinjia\u002FSkyAR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.11800)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fjiupinjia.github.io\u002Fskyar\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zal9Ues0aOQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fjiupinjia\u002FSkyAR\u002Fblob\u002Fmaster\u002Fcolab_demo.ipynb) | 18.01.2021 |\n| MusicXML Documentation | The goal of this notebook is to explore one of the magenta libraries for music | \u003Cul>\u003Cli>[Prakruti Joshi](https:\u002F\u002Fgithub.com\u002Fprakruti-joshi)\u003C\u002Fli> \u003Cli>[Falak Shah](https:\u002F\u002Ffalaktheoptimist.github.io\u002F)\u003C\u002Fli> \u003Cli>[Twisha Naik](https:\u002F\u002Fgithub.com\u002Ftwisha96)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>[magenta](https:\u002F\u002Fmagenta.tensorflow.org\u002F)\u003C\u002Fli>\u003Cli>[music theory](http:\u002F\u002Fmusictheoryblog.blogspot.com\u002F2008\u002F02\u002Flearn-music-theory.html)\u003C\u002Fli>\u003Cli>[musicXML](https:\u002F\u002Fwww.musicxml.com\u002Ffor-developers\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmagenta\u002Fmagenta-demos\u002Fblob\u002Fmaster\u002Fcolab-notebooks\u002FMusicXML_Document_Structure_Documentation.ipynb) | 08.01.2021 |\n| SVG VAE | A colab demo for the SVG VAE model | [Raphael Gontijo Lopes](https:\u002F\u002Fraphagl.com\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_3ad9d536e095.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV.2019.00802) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.02632)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fmagenta.tensorflow.org\u002Fsvg-vae)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmagenta\u002Fmagenta-demos\u002Fblob\u002Fmaster\u002Fcolab-notebooks\u002Fvae_svg_decoding.ipynb) | 08.01.2021 |\n| Neural Magic Eye | Learning to See and Understand the Scene Behind an Autostereogram | \u003Cul>\u003Cli>[Zhengxia Zou](http:\u002F\u002Fwww-personal.umich.edu\u002F~zzhengxi\u002F)\u003C\u002Fli> \u003Cli>[Tianyang Shi](https:\u002F\u002Fwww.shitianyang.tech\u002F)\u003C\u002Fli> \u003Cli>[Yi Yuan](https:\u002F\u002Fyiyuan1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhenwei Shi](http:\u002F\u002Flevir.buaa.edu.cn\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjiupinjia\u002Fneural-magic-eye?style=social)](https:\u002F\u002Fgithub.com\u002Fjiupinjia\u002Fneural-magic-eye) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.15692)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fjiupinjia.github.io\u002Fneuralmagiceye\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=Fkh7DEblqJ8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1f59dFLJ748i2TleE54RkbUZSMo9Hyx7l) | 01.01.2021 |\n| FGVC | Method first extracts and completes motion edges, and then uses them to guide piecewise-smooth flow completion with sharp edges | \u003Cul>\u003Cli>[Chen Gao](http:\u002F\u002Fchengao.vision\u002F)\u003C\u002Fli> \u003Cli>[Ayush Saraf](https:\u002F\u002Fgithub.com\u002Fayush29feb)\u003C\u002Fli> \u003Cli>[Johannes Kopf](https:\u002F\u002Fjohanneskopf.de\u002F)\u003C\u002Fli> \u003Cli>[Jia-Bin Huang](https:\u002F\u002Fjbhuang0604.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2af4fe162fb6.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-58610-2_42) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fvt-vl-lab\u002FFGVC?style=social)](https:\u002F\u002Fgithub.com\u002Fvt-vl-lab\u002FFGVC) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2009.01835)\u003C\u002Fli>\u003Cli>[project](http:\u002F\u002Fchengao.vision\u002FFGVC\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=CHHVPxHT7rc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1pb6FjWdwq_q445rG2NP0dubw7LKNUkqc) | 30.12.2020 |\n| VIBE | Video Inference for Body Pose and Shape Estimation, which makes use of an existing large-scale motion capture dataset together with unpaired, in-the-wild, 2D keypoint annotations | \u003Cul>\u003Cli>[Muhammed Kocabas](https:\u002F\u002Fps.is.mpg.de\u002Fperson\u002Fmkocabas)\u003C\u002Fli> \u003Cli>[Nikos Athanasiou](https:\u002F\u002Fgithub.com\u002Fathn-nik)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002Fperson\u002Fblack)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ed0b30996d1b.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00530) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmkocabas\u002FVIBE?style=social)](https:\u002F\u002Fgithub.com\u002Fmkocabas\u002FVIBE) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1912.05656)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcarlosedubarreto\u002Fvibe_win_install), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fakanazawa\u002Fhuman_dynamics), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMandyMo\u002Fpytorch_HMR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsoulslicer\u002FSTAF\u002Ftree\u002Fstaf)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002F3d-human-pose-estimation-on-3dpw?p=vibe-video-inference-for-human-body-pose-and)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3qhs5IRJ1LI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fw1biKeiQThY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FrIr-nX63dUA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfW0sIZfQcIs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8Qt0wA16kTo), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fxyo5gl5GLEI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FXNzgUhxKC38), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhErK0MamTY4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGfmm8uMfMq0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1dFfwxZ52MN86FA6uFNypMEdFShd2euQA) | 23.12.2020 |\n| SeFa | A closed-form approach for unsupervised latent semantic factorization in GANs | \u003Cul>\u003Cli>[Yujun Shen](https:\u002F\u002Fshenyujun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bolei Zhou](https:\u002F\u002Fboleizhou.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a10ddf728c4c.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00158) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgenforce\u002Fsefa?style=social)](https:\u002F\u002Fgithub.com\u002Fgenforce\u002Fsefa) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2007.06600)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgenforce.github.io\u002Fsefa\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=OFHW2WbXXIQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgenforce\u002Fsefa\u002Fblob\u002Fmaster\u002Fdocs\u002FSeFa.ipynb) | 06.12.2020 |\n| Stylized Neural Painting | An image-to-painting translation method that generates vivid and realistic painting artworks with controllable styles | \u003Cul>\u003Cli>[Zhengxia Zou](http:\u002F\u002Fwww-personal.umich.edu\u002F~zzhengxi\u002F)\u003C\u002Fli> \u003Cli>[Tianyang Shi](https:\u002F\u002Fwww.shitianyang.tech\u002F)\u003C\u002Fli> \u003Cli>[Yi Yuan](https:\u002F\u002Fyiyuan1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhenwei Shi](http:\u002F\u002Flevir.buaa.edu.cn\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_52b6cddbd876.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01543) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjiupinjia\u002Fstylized-neural-painting?style=social)](https:\u002F\u002Fgithub.com\u002Fjiupinjia\u002Fstylized-neural-painting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.08114)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fjiupinjia.github.io\u002Fneuralpainter\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=oerb-nwrXhk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1ch_41GtcQNQT1NLOA21vQJ_rQOjjv9D8) | 01.12.2020 |\n| BiT | Big Transfer: General Visual Representation Learning | \u003Cul>\u003Cli>[Alexander Kolesnikov](https:\u002F\u002Fgithub.com\u002Fakolesnikoff)\u003C\u002Fli> \u003Cli>[Lucas Beyer](http:\u002F\u002Flucasb.eyer.be)\u003C\u002Fli> \u003Cli>[Xiaohua Zhai](https:\u002F\u002Fgithub.com\u002Fxiaohuazhai)\u003C\u002Fli> \u003Cli>[Joan Puigcerver](https:\u002F\u002Fwww.jpuigcerver.net\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jessica Yung](https:\u002F\u002Fgithub.com\u002Fjessicayung)\u003C\u002Fli> \u003Cli>[Sylvain Gelly](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=m7LvuTkAAAAJ)\u003C\u002Fli> \u003Cli>[Neil Houlsby](https:\u002F\u002Fneilhoulsby.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_37c60b5e4977.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-58558-7_29) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fbig_transfer?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fbig_transfer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1912.11370), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.05237)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fgoogle\u002Fbit-50)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Freview-big-transfer-bit-general-visual-representation-learning-cb4bf8ed9732)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fk1GOF2jmX7c), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0iTgt5-SOsU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FX5Rhm__OxvA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Fbig_transfer\u002Fblob\u002Fmaster\u002Fcolabs\u002Fbig_transfer_tf2.ipynb) | 12.11.2020 |\n| LaSAFT | Latent Source Attentive Frequency Transformation for Conditioned Source Separation | [Woosung Choi](https:\u002F\u002Fws-choi.github.io\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_4563a7746cdb.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICASSP39728.2021.9413896) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fws-choi\u002FConditioned-Source-Separation-LaSAFT?style=social)](https:\u002F\u002Fgithub.com\u002Fws-choi\u002FConditioned-Source-Separation-LaSAFT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.11631)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fsigsep.github.io\u002Fdatasets\u002Fmusdb.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Flasaft.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fws-choi\u002FConditioned-Source-Separation-LaSAFT\u002Fblob\u002Fmaster\u002Fcolab_demo\u002FLaSAFT_with_GPoCM_Stella_Jang_Example.ipynb) | 01.11.2020 |\n| Lifespan Age Transformation Synthesis | Multi-domain image-to-image generative adversarial network architecture, whose learned latent space models a continuous bi-directional aging process | \u003Cul>\u003Cli>[Roy Or-El](https:\u002F\u002Fhomes.cs.washington.edu\u002F~royorel\u002F)\u003C\u002Fli> \u003Cli>[Soumyadip Sengupta](https:\u002F\u002Fhomes.cs.washington.edu\u002F~soumya91\u002F)\u003C\u002Fli> \u003Cli>[Ohad Fried](https:\u002F\u002Fwww.ohadf.com\u002F)\u003C\u002Fli> \u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli> \u003Cli>[Ira Kemelmacher-Shlizerman](https:\u002F\u002Fwww.irakemelmacher.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_083d7e30841c.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-58539-6_44) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Froyorel\u002FLifespan_Age_Transformation_Synthesis?style=social)](https:\u002F\u002Fgithub.com\u002Froyorel\u002FLifespan_Age_Transformation_Synthesis) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.09764)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Froyorel\u002FFFHQ-Aging-Dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fpix2pixHD), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstyle-based-gan-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgrail.cs.washington.edu\u002Fprojects\u002Flifespan_age_transformation_synthesis\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_jTFcjN2hBk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F9fulnt2_q_Y)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Froyorel\u002FLifespan_Age_Transformation_Synthesis\u002Fblob\u002Fmaster\u002FLATS_demo.ipynb) | 31.10.2020 |\n| IDInvert | In-domain GAN inversion approach, which not only faithfully reconstructs the input image but also ensures the inverted code to be semantically meaningful for editing | \u003Cul>\u003Cli>[Jiapeng Zhu](https:\u002F\u002Fgithub.com\u002Fzhujiapeng)\u003C\u002Fli> \u003Cli>[Yujun Shen](https:\u002F\u002Fshenyujun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Deli Zhao](https:\u002F\u002Fzhaodeli.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bolei Zhou](https:\u002F\u002Fboleizhou.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1ba54266a086.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-58520-4_35) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgenforce\u002Fidinvert?style=social)](https:\u002F\u002Fgithub.com\u002Fgenforce\u002Fidinvert) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.00049)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgenforce\u002Fidinvert_pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffyu\u002Flsun), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffyu\u002Flsun), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fai-innovation\u002Fin-domain-gan-inversion-for-anime-character-f6341d72e835)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgenforce.github.io\u002Fidinvert\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2qMw8sOsNg0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3v6NHrhuyFY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTVSJO9uNq7g)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgenforce\u002Fidinvert_pytorch\u002Fblob\u002Fmaster\u002Fdocs\u002FIdinvert.ipynb) | 22.10.2020 |\n| HiGAN | Semantic Hierarchy Emerges in Deep Generative Representations for Scene Synthesis | \u003Cul>\u003Cli>[Ceyuan Yang](https:\u002F\u002Fceyuan.me\u002F)\u003C\u002Fli> \u003Cli>[Yujun Shen](https:\u002F\u002Fshenyujun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bolei Zhou](https:\u002F\u002Fboleizhou.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_e9598715f8e3.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs11263-020-01429-5) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgenforce\u002Fhigan?style=social)](https:\u002F\u002Fgithub.com\u002Fgenforce\u002Fhigan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1911.09267), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1412.6856), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.10112)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgenforce.github.io\u002Fhigan\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=X5yWu2Jwjpg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgenforce\u002Fhigan\u002Fblob\u002Fmaster\u002Fdocs\u002FHiGAN_Bedroom.ipynb) | 14.10.2020 |\n| InterFaceGAN | Interpreting the Latent Space of GANs for Semantic Face Editing | \u003Cul>\u003Cli>[Yujun Shen](https:\u002F\u002Fshenyujun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jinjin Gu](https:\u002F\u002Fwww.jasongt.com\u002F)\u003C\u002Fli> \u003Cli>[Xiaoou Tang](https:\u002F\u002Fwww.ie.cuhk.edu.hk\u002Fpeople\u002Fxotang.shtml)\u003C\u002Fli> \u003Cli>[Bolei Zhou](https:\u002F\u002Fboleizhou.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_27a1ab666155.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00926) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgenforce\u002Finterfacegan?style=social)](https:\u002F\u002Fgithub.com\u002Fgenforce\u002Finterfacegan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.10786), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.09635), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1710.10196)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftkarras\u002Fprogressive_growing_of_gans), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgenforce.github.io\u002Finterfacegan\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=uoftpl3Bj6w)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgenforce\u002Finterfacegan\u002Fblob\u002Fmaster\u002Fdocs\u002FInterFaceGAN.ipynb) | 13.10.2020 |\n| Instance-aware Image Colorization | Novel deep learning framework to achieve instance-aware colorization | [Jheng-Wei Su](https:\u002F\u002Fgithub.com\u002Fericsujw) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0d2de5cb9566.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00799) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fericsujw\u002FInstColorization?style=social)](https:\u002F\u002Fgithub.com\u002Fericsujw\u002FInstColorization) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.10825)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fericsujw.github.io\u002FInstColorization\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=Zj1N4uE1ehk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fericsujw\u002FInstColorization\u002Fblob\u002Fmaster\u002FInstColorization.ipynb) | 30.08.2020 |\n| MnasNet | Automated mobile neural architecture search approach, which explicitly incorporate model latency into the main objective so that the search can identify a model that achieves a good trade-off between accuracy and latency | \u003Cul>\u003Cli>[Mingxing Tan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=6POeyBoAAAAJ)\u003C\u002Fli> \u003Cli>[Bo Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=HRDIoP0AAAAJ)\u003C\u002Fli> \u003Cli>[Ruoming Pang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=1fsmwB8AAAAJ)\u003C\u002Fli> \u003Cli>[Vijay Vasudevan](https:\u002F\u002Fvijay.vasu.org)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Mark Sandler](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=IcPc-OUAAAAJ)\u003C\u002Fli> \u003Cli>[Andrew Howard](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=_9l8vD8AAAAJ)\u003C\u002Fli> \u003Cli>[Quoc Le](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vfT6-XIAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_caae2795076f.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR.2019.00293) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Ftpu?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Ftpu\u002Ftree\u002Fmaster\u002Fmodels\u002Fofficial\u002Fmnasnet) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1807.11626)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAnjieCheng\u002FMnasNet-PyTorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fanalytics-vidhya\u002Fan-overview-on-mnasnet-platform-aware-neural-architecture-search-for-mobile-8a681d17a80c)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fdocs.pytorch.org\u002Fvision\u002Fmain\u002Fmodels\u002Fmnasnet.html)\u003C\u002Fli>\u003Cli>[tutorial](https:\u002F\u002Fcloud.google.com\u002Ftpu\u002Fdocs\u002Ftutorials\u002Fmnasnet)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F4uDZxefPd-I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F9hHKgPg4Wy0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdOwS37yZSew), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRkCa8OHSF9w), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F9kqfz0qPSW8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Ftpu\u002Fblob\u002Fmaster\u002Fmodels\u002Fofficial\u002Fmnasnet\u002Fmnasnet_example.ipynb) | 27.08.2020 |\n| MoCo | Momentum Contrast for unsupervised visual representation learning | \u003Cul>\u003Cli>[Kaiming He](https:\u002F\u002Fkaiminghe.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haoqi Fan](https:\u002F\u002Fhaoqifan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuxin Wu](https:\u002F\u002Fppwwyyxx.com\u002F)\u003C\u002Fli> \u003Cli>[Saining Xie](http:\u002F\u002Fsainingxie.com\u002F)\u003C\u002Fli> \u003Cli>[Ross Girshick](https:\u002F\u002Fwww.rossgirshick.info\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ff14d4f09568.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00975) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fmoco?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fmoco) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1911.05722), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.04297), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.02677)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fppwwyyxx\u002Fmoco.tensorflow)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FLvHwBQF14zs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F4VVGtYPM8JE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fo5Qh61dLDf0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fmoco\u002Fblob\u002Fcolab-notebook\u002Fcolab\u002Fmoco_cifar10_demo.ipynb) | 20.08.2020 |\n| CAPE | Learning to Dress 3D People in Generative Clothing | \u003Cul>\u003Cli>[Qianli Ma](https:\u002F\u002Fqianlim.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jinlong Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=HGt39SUAAAAJ)\u003C\u002Fli> \u003Cli>[Anurag Ranjan](https:\u002F\u002Fanuragranj.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sergi Pujades](https:\u002F\u002Fgithub.com\u002Fpujades)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Gerard Pons-Moll](https:\u002F\u002Fvirtualhumans.mpi-inf.mpg.de\u002F)\u003C\u002Fli> \u003Cli>[Siyu Tang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=BUDh_4wAAAAJ)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fc622cc59a15.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00650) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fqianlim\u002FCAPE?style=social)](https:\u002F\u002Fgithub.com\u002Fqianlim\u002FCAPE) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.13615), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1807.10267), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.02658)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcape.is.tue.mpg.de\u002Fdataset)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMPI-IS\u002Fmesh), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fanuragranj\u002Fcoma)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@mahyarfardinfar\u002Flearning-to-dress-3d-people-in-generative-clothing-486eb90136ff)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fcape.is.tue.mpg.de\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fe4W-hPFNwDE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FNOEA-Rtq6vM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1DCNo2OyyTNi1xDG-7j32FZQ9sBA6i9Ys) | 05.08.2020 |\n| Rewriting a Deep Generative Model | We ask if a deep network can be reprogrammed to follow different rules, by enabling a user to directly change the weights, instead of training with a data set | \u003Cul>\u003Cli>[David Bau](https:\u002F\u002Fpeople.csail.mit.edu\u002Fdavidbau\u002Fhome\u002F)\u003C\u002Fli> \u003Cli>[Steven Liu](http:\u002F\u002Fpeople.csail.mit.edu\u002Fstevenliu\u002F)\u003C\u002Fli> \u003Cli>[Tongzhou Wang](https:\u002F\u002Fssnl.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jun-Yan Zhu](https:\u002F\u002Fwww.cs.cmu.edu\u002F~junyanz\u002F)\u003C\u002Fli> \u003Cli>[Antonio Torralba](https:\u002F\u002Fgroups.csail.mit.edu\u002Fvision\u002Ftorralbalab\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_b5b78a288583.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-58452-8_21) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdavidbau\u002Frewriting?style=social)](https:\u002F\u002Fgithub.com\u002Fdavidbau\u002Frewriting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2007.15646), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1912.04958)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Frewriting.csail.mit.edu\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=i2_-zNqtEPk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Frewriting.csail.mit.edu\u002Fvideo\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdavidbau\u002Frewriting\u002Fblob\u002Fmaster\u002Fnotebooks\u002Frewriting-interface.ipynb) | 01.08.2020 |\n| SIREN | Implicit Neural Representations with Periodic Activation Functions | \u003Cul>\u003Cli>[Vincent Sitzmann](https:\u002F\u002Fvsitzmann.github.io\u002F)\u003C\u002Fli> \u003Cli>[Julien Martel](http:\u002F\u002Fweb.stanford.edu\u002F~jnmartel\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fvsitzmann\u002Fsiren?style=social)](https:\u002F\u002Fgithub.com\u002Fvsitzmann\u002Fsiren) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.09661)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fdrive.google.com\u002Fdrive\u002Ffolders\u002F1_iq__37-hw7FJOEUK1tX7mdp8SKB368K)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2020\u002Fhash\u002F53c04118df112c13a8c34b38343b9c10-Abstract.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fvsitzmann.github.io\u002Fsiren\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=Q2fLWGBeaiI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fvsitzmann\u002Fsiren\u002Fblob\u002Fmaster\u002Fexplore_siren.ipynb) | 25.06.2020 |\n| 3D Photo Inpainting | Method for converting a single RGB-D input image into a 3D photo, i.e., a multi-layer representation for novel view synthesis that contains hallucinated color and depth structures in regions occluded in the original view | \u003Cul>\u003Cli>[Meng-Li Shih](https:\u002F\u002Fshihmengli.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shih-Yang Su](https:\u002F\u002Flemonatsu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Johannes Kopf](https:\u002F\u002Fjohanneskopf.de\u002F)\u003C\u002Fli> \u003Cli>[Jia-Bin Huang](https:\u002F\u002Fjbhuang0604.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_77288fc3bc90.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00805) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fvt-vl-lab\u002F3d-photo-inpainting?style=social)](https:\u002F\u002Fgithub.com\u002Fvt-vl-lab\u002F3d-photo-inpainting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.04727)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fshihmengli.github.io\u002F3D-Photo-Inpainting\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1706ToQrkIZshRSJSHvZ1RuCiM__YX3Bz) | 04.05.2020 |\n| Audio-driven Talking Face Video Generation with Learning-based Personalized Head Pose | Deep neural network model that takes an audio signal A of a source person and a very short video V of a target person as input, and outputs a synthesized high-quality talking face video with personalized head pose (making use of the visual information in V), expression and lip synchronization (by considering both A and V) | \u003Cul>\u003Cli>[Ran Yi](https:\u002F\u002Fyiranran.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zipeng Ye](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=faIXk9EAAAAJ)\u003C\u002Fli> \u003Cli>[Juyong Zhang](https:\u002F\u002Fjuyong.github.io\u002Findex.html)\u003C\u002Fli> \u003Cli>[Hujun Bao](https:\u002F\u002Fieeexplore.ieee.org\u002Fauthor\u002F37271755400)\u003C\u002Fli> \u003Cli>[Yong-Jin Liu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=GNDtwWQAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1c16c70541ee.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTMM.2022.3207606) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyiranran\u002FAudio-driven-TalkingFace-HeadPose?style=social)](https:\u002F\u002Fgithub.com\u002Fyiranran\u002FAudio-driven-TalkingFace-HeadPose) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2002.10137)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJuyong\u002F3DFace), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FDeep3DFaceReconstruction), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTEShHzqoTwk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1gqcqTSAGAyj48n0fmApvSPG_43BzKP37) | 29.04.2020 |\n| Motion Supervised co-part Segmentation | A self-supervised deep learning method for co-part segmentation | \u003Cul>\u003Cli>[Aliaksandr Siarohin](https:\u002F\u002Faliaksandrsiarohin.github.io\u002Faliaksandr-siarohin-website\u002F)\u003C\u002Fli> \u003Cli>[Subhankar Roy](https:\u002F\u002Fgithub.com\u002Froysubhankar)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_4664349ec5ee.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICPR48806.2021.9412520) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAliaksandrSiarohin\u002Fmotion-cosegmentation?style=social)](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Fmotion-cosegmentation) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F2004.03234)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Fvideo-preprocessing)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=RJ4Nj1wV5iA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAliaksandrSiarohin\u002Fmotion-cosegmentation\u002Fblob\u002Fmaster\u002Fpart_swap.ipynb) | 07.04.2020 |\n| Onsets and Frames | Onsets and Frames is an automatic music transcription framework with piano and drums models | \u003Cul>\u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli> \u003Cli>[Erich Elsen](https:\u002F\u002Fgithub.com\u002Fekelsen)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagenta\u002Fmagenta?style=social)](https:\u002F\u002Fgithub.com\u002Fmagenta\u002Fmagenta\u002Ftree\u002Fmain\u002Fmagenta\u002Fmodels\u002Fonsets_frames_transcription) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1710.11153), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.12247), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.00188)\u003C\u002Fli>\u003Cli>[blog post](http:\u002F\u002Fg.co\u002Fmagenta\u002Fonsets-frames)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fg.co\u002Fmagenta\u002Fmaestro-wave2midi2wave), [data](https:\u002F\u002Fmagenta.tensorflow.org\u002Fdatasets\u002Fe-gmd)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Fmagenta\u002Fonsets_frames_transcription\u002Fonsets_frames_transcription.ipynb) | 02.04.2020 |\n| FBA Matting | Low-cost modification to alpha matting networks to also predict the foreground and background colours | \u003Cul>\u003Cli>[Marco Forte](https:\u002F\u002Fgithub.com\u002FMarcoForte)\u003C\u002Fli> \u003Cli>[François Pitié](https:\u002F\u002Ffrancois.pitie.net\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FMarcoForte\u002FFBA_Matting?style=social)](https:\u002F\u002Fgithub.com\u002FMarcoForte\u002FFBA_Matting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.07711)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMarcoForte\u002Fclosed-form-matting)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fleonelhs\u002FFBA-Matting)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fimage-matting-on-composition-1k?p=f-b-alpha-matting)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Ut2szLBTxPejGHt_GYUkua21yUVWseOE) | 19.03.2020 |\n| BERT score | An automatic evaluation metric for text generation | [Tianyi Zhang](https:\u002F\u002Ftiiiger.github.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTiiiger\u002Fbert_score?style=social)](https:\u002F\u002Fgithub.com\u002FTiiiger\u002Fbert_score) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.09675)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fbert-score\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FTiiiger\u002Fbert_score\u002Fblob\u002Fmaster\u002Fexample\u002FDemo.ipynb) | 05.03.2020 |\n| Deep Image Prior | Structure of a generator network is sufficient to capture a great deal of low-level image statistics prior to any learning | \u003Cul>\u003Cli>[Dmitry Ulyanov](https:\u002F\u002Fdmitryulyanov.github.io\u002Fabout)\u003C\u002Fli> \u003Cli>[Andrea Vedaldi](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vedaldi\u002F)\u003C\u002Fli> \u003Cli>[Victor Lempitsky](http:\u002F\u002Fsites.skoltech.ru\u002Fcompvision\u002Fmembers\u002Fvilem\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_443e7051c277.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs11263-020-01303-4) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FDmitryUlyanov\u002Fdeep-image-prior?style=social)](https:\u002F\u002Fgithub.com\u002FDmitryUlyanov\u002Fdeep-image-prior) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.10925)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fdmitryulyanov.github.io\u002Fdeep_image_prior)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002F7gls3j\u002Fr_deep_image_prior_deep_superresolution\u002F)\u003C\u002Fli>\u003Cli>[supmat](https:\u002F\u002Fbox.skoltech.ru\u002Findex.php\u002Fs\u002Fib52BOoV58ztuPM)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FDeep_image_prior)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-g1NsTuP1_I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_BPJFFkxSbw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_WcGdXPdfjo), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfCneY-7zFXE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUYfPet5w_34), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIxIMvwkUsiQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3X5zfV5eQlY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FLzrUQtH43fY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FDmitryUlyanov\u002Fdeep-image-prior\u002Fblob\u002Fmaster\u002Factivation_maximization.ipynb) | 30.10.2019 |\n| ProxylessNAS | Directly learn the architectures for large-scale target tasks and target hardware platforms | \u003Cul>\u003Cli>[Han Cai](https:\u002F\u002Fhan-cai.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ligeng Zhu](https:\u002F\u002Flzhu.me\u002F)\u003C\u002Fli> \u003Cli>[Song Han](https:\u002F\u002Fhanlab.mit.edu\u002Fsonghan)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmit-han-lab\u002Fproxylessnas?style=social)](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fproxylessnas) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.00332), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.03494), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.08886)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Fpaper-proxylessnas-direct-neural-architecture-search-on-target-task-image-classification-73c35ebd8aed)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fpytorch.org\u002Fhub\u002Fpytorch_vision_proxylessnas\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Fa3a1xy\u002Fr_proxylessnas_direct_neural_architecture_search\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FX0aZmppnO1s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F6AeCIJH9eGI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fpytorch\u002Fpytorch.github.io\u002Fblob\u002Fmaster\u002Fassets\u002Fhub\u002Fpytorch_vision_proxylessnas.ipynb) | 29.10.2019 |\n| Generating Piano Music with Transformer | This Colab notebook lets you play with pretrained Transformer models for piano music generation, based on the Music Transformer | \u003Cul>\u003Cli>[Ian Simon](https:\u002F\u002Fgithub.com\u002Fiansimon)\u003C\u002Fli> \u003Cli>[Anna Huang](https:\u002F\u002Fgithub.com\u002Fczhuang)\u003C\u002Fli> \u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli> \u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.03762), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.04281)\u003C\u002Fli>\u003Cli>[blog post](http:\u002F\u002Fg.co\u002Fmagenta\u002Fmusic-transformer)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Fmagenta\u002Fpiano_transformer\u002Fpiano_transformer.ipynb) | 16.09.2019 |\n| SSGAN | Self-Supervised GANs via Auxiliary Rotation Loss | \u003Cul>\u003Cli>[Ting Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=KoXUMbsAAAAJ)\u003C\u002Fli> \u003Cli>[Xiaohua Zhai](https:\u002F\u002Fsites.google.com\u002Fview\u002Fxzhai)\u003C\u002Fli> \u003Cli>[Marvin Ritter](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=arcf5FgAAAAJ)\u003C\u002Fli> \u003Cli>[Mario Lučić](https:\u002F\u002Flucic.ai\u002F)\u003C\u002Fli> \u003Cli>[Neil Houlsby](https:\u002F\u002Fneilhoulsby.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_55478530ab65.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR.2019.01243) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fcompare_gan?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fcompare_gan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.11212), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.10337), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1807.04720), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1806.00035), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.04874), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.10340), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.01365), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.11212), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.02271)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fmodels\u002Fgoogle\u002Fcompare-gan)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science\u002Fself-supervised-gans-using-auxiliary-rotation-loss-60d8a929b556)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-oJWFcexolY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fcompare_gan\u002Fblob\u002Fmaster\u002Fcolabs\u002Fssgan_demo.ipynb) | 20.06.2019 |\n| S3GAN | High-Fidelity Image Generation With Fewer Labels | \u003Cul>\u003Cli>[Mario Lucic](https:\u002F\u002Fresearch.google\u002Fpeople\u002Fmariolucic\u002F)\u003C\u002Fli> \u003Cli>[Michael Tschannen](https:\u002F\u002Fmitscha.github.io)\u003C\u002Fli> \u003Cli>[Marvin Ritter](https:\u002F\u002Fgithub.com\u002Fmritter0)\u003C\u002Fli> \u003Cli>[Xiaohua Zhai](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=sEkgK04AAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Olivier Bachem](https:\u002F\u002Fresearch.google\u002Fpeople\u002Folivierbachem\u002F)\u003C\u002Fli> \u003Cli>[Sylvain Gelly](https:\u002F\u002Fwww.chessprogramming.org\u002FSylvain_Gelly)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fcompare_gan?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fcompare_gan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.02271)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fcode\u002Fkerneler\u002Fgenerating-images-with-little-data-using-s3gan), [\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fmodels\u002Fgoogle\u002Fcompare-gan\u002FTensorFlow1\u002Fs3gan-20-128x128\u002F1)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fhub\u002Ftutorials\u002Fs3gan_generation_with_tf_hub)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fcompare_gan\u002Fblob\u002Fmaster\u002Fcolabs\u002Fs3gan_demo.ipynb) | 10.06.2019 |\n| HMR | End-to-end framework for reconstructing a full 3D mesh of a human body from a single RGB image | \u003Cul>\u003Cli>[Angjoo Kanazawa](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~kanazawa\u002F)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002Fperson\u002Fblack)\u003C\u002Fli> \u003Cli>[David Jacobs](https:\u002F\u002Fwww.cs.umd.edu\u002F~djacobs\u002F)\u003C\u002Fli> \u003Cli>[Jitendra Malik](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~malik\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_abd81e31ab98.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR.2018.00744) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fakanazawa\u002Fhmr?style=social)](https:\u002F\u002Fgithub.com\u002Fakanazawa\u002Fhmr) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1712.06584)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fdawars\u002Fhmr\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmattloper\u002Fchumpy), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCMU-Perceptual-Computing-Lab\u002Fopenpose), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMandyMo\u002Fpytorch_HMR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flayumi\u002Fhmr), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frussoale\u002Fhmr2.0)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fakanazawa.github.io\u002Fhmr\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbmMV9aJKa-c)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FDene33\u002Fvideo_to_bvh\u002Fblob\u002Fmaster\u002Fvideo_to_bvh.ipynb) | 15.03.2019 |\n| GANSynth | This notebook is a demo GANSynth, which generates audio with Generative Adversarial Networks | [Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagenta\u002Fmagenta?style=social)](https:\u002F\u002Fgithub.com\u002Fmagenta\u002Fmagenta\u002Ftree\u002Fmain\u002Fmagenta\u002Fmodels\u002Fgansynth) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.08710), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.11096)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fstorage.googleapis.com\u002Fmagentadata\u002Fpapers\u002Fgansynth\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Fmagenta\u002Fgansynth\u002Fgansynth_demo.ipynb) | 25.02.2019 |\n| AmoebaNet | Regularized Evolution for Image Classifier Architecture Search | \u003Cul>\u003Cli>[Esteban Real](https:\u002F\u002Fwww.estebanreal.com\u002F)\u003C\u002Fli> \u003Cli>[Alok Aggarwal](https:\u002F\u002Fscryai.com\u002Fabout-us\u002Falok-aggarwal\u002F)\u003C\u002Fli> \u003Cli>[Yanping Huang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=uEtBQScAAAAJ)\u003C\u002Fli> \u003Cli>[Quoc Le](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FQuoc_V._Le)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2b2991845958.png)](https:\u002F\u002Fdoi.org\u002F10.1609\u002Faaai.v33i01.33014780) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fgoogle-research?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fgoogle-research\u002Ftree\u002Fmaster\u002Fevolution\u002Fregularized_evolution_algorithm) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.01548)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Freading-amoebanet-regularized-evolution-for-image-classifier-architecture-search-image-278f5c077a4a)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fg557Nhg-f-k), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FOROmOKlcRxs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAYYhrfRPyww)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Fgoogle-research\u002Fblob\u002Fmaster\u002Fevolution\u002Fregularized_evolution_algorithm\u002Fregularized_evolution.ipynb) | 25.10.2018 |\n| Latent Constraints | Conditional Generation from Unconditional Generative Models | \u003Cul>\u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli> \u003Cli>[Matthew Hoffman](http:\u002F\u002Fmatthewdhoffman.com\u002F)\u003C\u002Fli> \u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.05772)\u003C\u002Fli>\u003Cli>[data](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FCelebA.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Flatent_constraints\u002Flatentconstraints.ipynb) | 27.11.2017 |\n| Performance RNN | This notebook shows you how to generate new performed compositions from a trained model | \u003Cul>\u003Cli>[Ian Simon](https:\u002F\u002Fgithub.com\u002Fiansimon)\u003C\u002Fli> \u003Cli>[Sageev Oore](https:\u002F\u002Fgithub.com\u002Fosageev)\u003C\u002Fli> \u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagenta\u002Fmagenta?style=social)](https:\u002F\u002Fgithub.com\u002Fmagenta\u002Fmagenta\u002Ftree\u002Fmaster\u002Fmagenta\u002Fmodels\u002Fperformance_rnn) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fmagenta.tensorflow.org\u002Fperformance-rnn)\u003C\u002Fli>\u003Cli>[data](http:\u002F\u002Fwww.piano-e-competition.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Fmagenta\u002Fperformance_rnn\u002Fperformance_rnn.ipynb) | 11.07.2017 |\n| NSynth | This colab notebook has everything you need to upload your own sounds and use NSynth models to reconstruct and interpolate between them | \u003Cul>\u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli> \u003Cli>[Cinjon Resnick](https:\u002F\u002Fgithub.com\u002Fcinjon)\u003C\u002Fli> \u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Sander Dieleman](https:\u002F\u002Fbenanne.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Karen Simonyan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=L7lMQkQAAAAJ)\u003C\u002Fli> \u003Cli>[Mohammad Norouzi](https:\u002F\u002Fnorouzi.github.io\u002F)\u003C\u002Fli> \u003Cli>[Douglas Eck](https:\u002F\u002Fgithub.com\u002Fdouglaseck)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Fmagenta?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fmagenta\u002Ftree\u002Fmaster\u002Fmagenta\u002Fmodels\u002Fnsynth) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1704.01279)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fmagenta.tensorflow.org\u002Fnsynth)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fmagenta.tensorflow.org\u002Fdatasets\u002Fnsynth)\u003C\u002Fli>\u003Cli>[tutorial](https:\u002F\u002Fmagenta.tensorflow.org\u002Fnsynth-fastgen)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=AaALLWQmCdI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=BOoSy-Pg8is)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Fmagenta\u002Fnsynth\u002Fnsynth.ipynb) | 06.04.2017 |\n\n\u003C\u002Fdetails>\n\n## Tutorials\n\u003Cdetails>\n\u003Csummary>TUTORIALS\u003C\u002Fsummary>\n\n| name | description | authors | links | colaboratory | update |\n|------|-------------|:--------|:------|:------------:|:------:|\n| Kornia | Library is composed by a subset of packages containing operators that can be inserted within neural networks to train models to perform image transformations, epipolar geometry, depth estimation, and low-level image processing such as filtering and edge detection that operate directly on tensors | \u003Cul>\u003Cli>[Edgar Riba](https:\u002F\u002Fgithub.com\u002Fedgarriba)\u003C\u002Fli> \u003Cli>[Dmytro Mishkin](https:\u002F\u002Fdmytro.ai\u002F)\u003C\u002Fli> \u003Cli>[Daniel Ponsa](https:\u002F\u002Fgithub.com\u002FDanielPonsa)\u003C\u002Fli> \u003Cli>[Ethan Rublee](https:\u002F\u002Fgithub.com\u002Fethanrublee)\u003C\u002Fli> \u003Cli>[Gary Bradski](https:\u002F\u002Fgithub.com\u002Fgarybradski)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_7f982a5488a1.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FWACV45572.2020.9093363) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fkornia\u002Fkornia?style=social)](https:\u002F\u002Fgithub.com\u002Fkornia\u002Fkornia) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.02190)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fopencv.org\u002Fkornia-an-open-source-differentiable-computer-vision-library-for-pytorch\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FHfnywwpBnD)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fkornia.readthedocs.io\u002Fen\u002Flatest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fkornia\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fslack.svg\" alt=\"slack\" height=20\u002F>](https:\u002F\u002Fjoin.slack.com\u002Ft\u002Fkornia\u002Fshared_invite\u002Fzt-csobk21g-2AQRi~X9Uu6PLMuUZdvfjA)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002Fkornia_foss)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fkornia.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fchannel\u002FUCI1SE1Ij2Fast5BSKxoa7Ag), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3RmCYFhwclE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAAZa-mXjYF0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fkornia\u002Fkornia\u002Fblob\u002Fmaster\u002Fexamples\u002Faugmentation\u002Fkornia_augmentation.ipynb) | 28.11.2025 |\n| LM Evaluation Harness | Framework for few-shot evaluation of language models. | [Lintang Sutawika](https:\u002F\u002Flintang.sutawika.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FEleutherAI\u002Flm-evaluation-harness?style=social)](https:\u002F\u002Fgithub.com\u002FEleutherAI\u002Flm-evaluation-harness) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.14165)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002Feleutherai)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAutoGPTQ\u002FAutoGPTQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEleutherAI\u002Fgpt-neox), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FMegatron-DeepSpeed), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvllm-project\u002Fvllm)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.eleuther.ai\u002Fprojects\u002Flarge-language-model-evaluation)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FEleutherAI\u002Flm-evaluation-harness\u002Fblob\u002Fmain\u002Fexamples\u002Flm-eval-overview.ipynb) | 26.11.2025 |\n| Magenta RT | An open-weights live music model that allows you to interactively create, control and perform music in the moment | [Chris Donahue](https:\u002F\u002Fchrisdonahue.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagenta\u002Fmagenta-realtime?style=social)](https:\u002F\u002Fgithub.com\u002Fmagenta\u002Fmagenta-realtime) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.03312), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.12415), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.01917)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fmagenta.withgoogle.com\u002Fmagenta-realtime)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fai.google.dev\u002Fgemini-api\u002Fdocs\u002Fmusic-generation)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fgoogle\u002Fmagenta-realtime)\u003C\u002Fli>\u003Cli>[labs](https:\u002F\u002Flabs.google\u002Ffx\u002Ftools\u002Fmusic-fx-dj\u002Funsupported-country)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science-in-your-pocket\u002Fgoogle-magenta-realtime-ai-can-now-generate-songs-0cb3dbe01a00)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fmagenta\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FSVTuEdeepVs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAe1Kz2zmh9M), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F6gVvsv3Va3s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FEg91A8sSWUM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3zkqFCaY3IQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmagenta\u002Fmagenta-realtime\u002Fblob\u002Fmain\u002Fnotebooks\u002FMagenta_RT_Demo.ipynb) | 26.11.2025 |\n| SHAP | SHapley Additive exPlanations is a game theoretic approach to explain the output of any machine learning model | \u003Cul>\u003Cli>[Scott Lundberg](https:\u002F\u002Fscottlundberg.com\u002F)\u003C\u002Fli> \u003Cli>[Su-In Lee](https:\u002F\u002Fwww.cs.washington.edu\u002Fpeople\u002Ffaculty\u002Fsu-in-lee\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a49ddcdcc789.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs42256-019-0138-9) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fshap\u002Fshap?style=social)](https:\u002F\u002Fgithub.com\u002Fshap\u002Fshap) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.13972), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1704.02685), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1703.01365), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.03825)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fshap.readthedocs.io\u002Fen\u002Flatest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2017\u002Fhash\u002F8a20a8621978632d76c43dfd28b67767-Abstract.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fshap.readthedocs.io\u002Fen\u002Flatest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVB9uV-x0gtg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fd4PPMpdUCz8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fwjd1G5bu_TY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fd6PsRiEKTb8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3BVU6nrfk4o), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FLsV-00K9W5I)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fshap\u002Fshap\u002Fblob\u002Fmaster\u002Fnotebooks\u002Foverviews\u002FAn%20introduction%20to%20explainable%20AI%20with%20Shapley%20values.ipynb#scrollTo=EAn8TeTgyA0Nhttps:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fshap\u002Fshap\u002Fblob\u002Fmaster\u002Fnotebooks\u002Foverviews\u002FAn%20introduction%20to%20explainable%20AI%20with%20Shapley%20values.ipynb) | 20.11.2025 |\n| Nano Banana | An image generation and editing model powered by generative artificial intelligence and developed by Google DeepMind | [Guillaume Vernade](https:\u002F\u002Fgithub.com\u002FGiom-V) | \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fblog.google\u002Ftechnology\u002Fai\u002Fnano-banana-pro\u002Fv)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fmodels\u002Fgemini-image\u002Fpro\u002F), [\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fmodels\u002Fgemini-image\u002Fflash\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fai.google.dev\u002Fgemini-api\u002Fdocs\u002Fimage-generation)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fgoogle-cloud\u002Fmy-experience-using-the-new-gemini-2-5-flash-image-8fbf79f00d76)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Faistudio.google.com\u002Fmodels\u002Fgemini-2-5-flash-image)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FNano_Banana)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8_GgeASwHwQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjtQiCJXOvdg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRJ2eqkk_JxI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIjWWnDZlezI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fm8Ve7hNl9P0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FexWEkRHmhKU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5PiOEbnBDBs)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-gemini\u002Fcookbook\u002Fblob\u002Fmain\u002Fquickstarts\u002FGet_Started_Nano_Banana.ipynb) | 20.11.2025 |\n| NeMo | A conversational AI toolkit built for researchers working on automatic speech recognition, natural language processing, and text-to-speech synthesis | \u003Cul>\u003Cli>[Oleksii Kuchaiev](http:\u002F\u002Fkuchaev.com\u002F)\u003C\u002Fli> \u003Cli>[Jason Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=V28bxDwAAAAJ)\u003C\u002Fli> \u003Cli>[Chip Huyen](https:\u002F\u002Fhuyenchip.com\u002F)\u003C\u002Fli> \u003Cli>[Oleksii Hrinchuk](https:\u002F\u002Fgithub.com\u002FAlexGrinch)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ryan Leary](https:\u002F\u002Fgithub.com\u002Fryanleary)\u003C\u002Fli> \u003Cli>[Boris Ginsburg](https:\u002F\u002Fgithub.com\u002Fborisgin)\u003C\u002Fli> \u003Cli>[Samuel Kriman](https:\u002F\u002Fgithub.com\u002Fsam1373)\u003C\u002Fli> \u003Cli>[Stanislav Beliaev](https:\u002F\u002Fgithub.com\u002Fstasbel)\u003C\u002Fli> \u003Cli>[Vitaly Lavrukhin](https:\u002F\u002Fgithub.com\u002Fvsl9)\u003C\u002Fli> \u003Cli>[Jack Cook](https:\u002F\u002Fjackcook.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNVIDIA\u002FNeMo?style=social)](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNeMo) \u003Cul>\u003Cli>[project](https:\u002F\u002Fdocs.nvidia.com\u002Fdeeplearning\u002Fnemo\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FwBgpMf_KQVw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FNVIDIA\u002FNeMo\u002Fblob\u002Fmaster\u002Ftutorials\u002F00_NeMo_Primer.ipynb) | 17.11.2025 |\n| PyTerrier | A Python framework for performing information retrieval experiments | \u003Cul>\u003Cli>[Craig Macdonald](https:\u002F\u002Fwww.dcs.gla.ac.uk\u002F~craigm\u002F)\u003C\u002Fli> \u003Cli>[Nicola Tonellotto](https:\u002F\u002Fgithub.com\u002Ftonellotto)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_346a1cde8fcc.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3459637.3482013) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fterrier-org\u002Fpyterrier?style=social)](https:\u002F\u002Fgithub.com\u002Fterrier-org\u002Fpyterrier) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2007.14271)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fpyterrier.readthedocs.io)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrier-org\u002Fecir2021tutorial), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_ance), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_colbert), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_pisa), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_t5), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_doc2query), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_deepct)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fterrier-org\u002Fpyterrier\u002Fblob\u002Fmaster\u002Fexamples\u002Fnotebooks\u002Fnon_en_retrieval.ipynb) | 13.11.2025 |\n| Transfer learning and fine-tuning | You will learn how to classify images of cats and dogs by using transfer learning from a pre-trained network | [François Chollet](https:\u002F\u002Ffchollet.com\u002F) | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Ftask\u002Ftransfer-learning)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Ftutorials\u002Fimages\u002Ftransfer_learning)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FTransfer_learning)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fdocs\u002Fblob\u002Fmaster\u002Fsite\u002Fen\u002Ftutorials\u002Fimages\u002Ftransfer_learning.ipynb) | 11.11.2025 |\n| Datasets | A Community Library for Natural Language Processing | \u003Cul>\u003Cli>[Quentin Lhoest](https:\u002F\u002Fgithub.com\u002Flhoestq)\u003C\u002Fli> \u003Cli>[Albert Villanova](https:\u002F\u002Falbertvillanova.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yacine Jernite](https:\u002F\u002Fyjernite.github.io\u002F)\u003C\u002Fli> \u003Cli>[Abhishek Thakur](https:\u002F\u002Fgithub.com\u002Fabhishekkrthakur)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Patrick von Platen](https:\u002F\u002Fgithub.com\u002Fpatrickvonplaten)\u003C\u002Fli> \u003Cli>[Suraj Patil](https:\u002F\u002Fgithub.com\u002Fpatil-suraj)\u003C\u002Fli> \u003Cli>[Julien Chaumond](https:\u002F\u002Fgithub.com\u002Fjulien-c)\u003C\u002Fli> \u003Cli>[Mariama Dramé](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=0pwfXH0AAAAJ)\u003C\u002Fli> \u003Cli>[Julien Plu](https:\u002F\u002Fjplu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Lewis Tunstall](https:\u002F\u002Flewtun.github.io\u002Fblog\u002F)\u003C\u002Fli> \u003Cli>[Joe Davison](https:\u002F\u002Fjoeddav.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mario Šaško](https:\u002F\u002Fgithub.com\u002Fmariosasko)\u003C\u002Fli> \u003Cli>[Gunjan Chhablani](https:\u002F\u002Fgchhablani.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bhavitvya Malik](https:\u002F\u002Fgithub.com\u002Fbhavitvyamalik)\u003C\u002Fli> \u003Cli>[Simon Brandeis](https:\u002F\u002Fgithub.com\u002FSBrandeis)\u003C\u002Fli> \u003Cli>[Teven Le Scao](https:\u002F\u002Fgithub.com\u002FTevenLeScao)\u003C\u002Fli> \u003Cli>[Victor Sanh](https:\u002F\u002Fgithub.com\u002FVictorSanh)\u003C\u002Fli> \u003Cli>[Canwen Xu](https:\u002F\u002Fwww.canwenxu.net\u002F)\u003C\u002Fli> \u003Cli>[Nicolas Patry](https:\u002F\u002Fgithub.com\u002FNarsil)\u003C\u002Fli> \u003Cli>[Angelina McMillan-Major](https:\u002F\u002Fgithub.com\u002Fmcmillanmajora)\u003C\u002Fli> \u003Cli>[Philipp Schmid](https:\u002F\u002Fwww.philschmid.de\u002F)\u003C\u002Fli> \u003Cli>[Sylvain Gugger](https:\u002F\u002Fgithub.com\u002Fsgugger)\u003C\u002Fli> \u003Cli>[Clément Delangue](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=bRMboT8AAAAJ)\u003C\u002Fli> \u003Cli>[Théo Matussière](https:\u002F\u002Ftheo.matussie.re\u002F)\u003C\u002Fli> \u003Cli>[Lysandre Debut](http:\u002F\u002Flysand.re\u002F)\u003C\u002Fli> \u003Cli>[Stas Bekman](https:\u002F\u002Fstasosphere.com\u002Fmachine-learning\u002F)\u003C\u002Fli> \u003Cli>[Pierric Cistac](https:\u002F\u002Fgithub.com\u002FPierrci)\u003C\u002Fli> \u003Cli>[Thibault Goehringer](https:\u002F\u002Fgithub.com\u002Fbeurkinger)\u003C\u002Fli> \u003Cli>[Victor Mustar](https:\u002F\u002Fgithub.com\u002Fgary149)\u003C\u002Fli> \u003Cli>[François Lagunas](https:\u002F\u002Fgithub.com\u002Fmadlag)\u003C\u002Fli> \u003Cli>[Alexander Rush](https:\u002F\u002Frush-nlp.com\u002F)\u003C\u002Fli> \u003Cli>[Thomas Wolf](https:\u002F\u002Fthomwolf.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhuggingface\u002Fdatasets?style=social)](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fdatasets) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.02846)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fdatasets)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fcode\u002Fnbroad\u002Fintro-to-hugging-face-datasets)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fdatasets\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuaIJ96syPnM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fhuggingface\u002Fnotebooks\u002Fblob\u002Fmain\u002Fdatasets_doc\u002Fen\u002Fquickstart.ipynb) | 10.11.2025 |\n| Agent Starter Pack | Collection of production-ready Generative AI Agent templates built for Google Cloud | [Kristopher Overholt](https:\u002F\u002Fgithub.com\u002Fkoverholt) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FGoogleCloudPlatform\u002Fagent-starter-pack?style=social)](https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fagent-starter-pack) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fgoogle-cloud\u002Fagentic-rag-with-bigquery-dataframes-and-agent-starter-pack-743553adf997), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fgoogle-cloud\u002Fgenai-app-starter-pack-now-with-rag-pattern-vertex-ai-search-fb81bf61bcd5)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fagent-starter-pack\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fgooglecloud\u002Fcomments\u002F1j49uz7\u002Fagent_starter_pack_build_deploy_genai_agents_on\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjHt-ZVD660g), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FkwRG7cnqSu0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=yIRIT_EtALs&t=235s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLIivdWyY5sqLRCzKJyixrIDPQKwU6XHpn)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGoogleCloudPlatform\u002Fgenerative-ai\u002Fblob\u002Fmain\u002Fgemini\u002Fagent-engine\u002Fintro_agent_engine.ipynb) | 06.11.2025 |\n| Google Cloud Text-to-Speech | Enables easy integration of Google text recognition technologies into developer applications | \u003Cul>\u003Cli>[Holt Skinner](https:\u002F\u002Fgithub.com\u002Fholtskinner)\u003C\u002Fli> \u003Cli>[Ivan Nardini](https:\u002F\u002Fgithub.com\u002Finardini)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogleapis\u002Fgoogle-cloud-python?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogleapis\u002Fgoogle-cloud-python\u002Ftree\u002Fmain\u002Fpackages\u002Fgoogle-cloud-texttospeech) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fcloud.google.com\u002Fpython\u002Fdocs\u002Freference\u002Ftexttospeech\u002Flatest\u002Fsummary_overview)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgoogle-cloud-texttospeech\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fcloud.google.com\u002Ftext-to-speech)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGVPWz-nhJhg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fwzp9dfVpeeg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FNaAB9ogyB-U), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fz8g3XM16eRM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDCsVRs1-Hn8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FSPcFViKU_xU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FlKra6E_tp5U), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdOlV_oD_dr8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGoogleCloudPlatform\u002Fgenerative-ai\u002Fblob\u002Fmain\u002Faudio\u002Fspeech\u002Fgetting-started\u002Fget_started_with_chirp_3_hd_voices.ipynb) | 06.11.2025 |\n| Imagen 4 | Text-to-image model, with photorealistic images, near real-time speed, and sharper clarity | [Katie Nguyen](https:\u002F\u002Fgithub.com\u002Fkatiemn) | \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fdevelopers.googleblog.com\u002Fen\u002Fimagen-4-now-available-in-the-gemini-api-and-google-ai-studio\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fmodels\u002Fimagen\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgoogle-genai\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-f4I1_27SwI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAzwg982tpvM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGoogleCloudPlatform\u002Fgenerative-ai\u002Fblob\u002Fmain\u002Fvision\u002Fgetting-started\u002Fimagen4_image_generation.ipynb) | 06.11.2025 |\n| Lyria 2 | Delivers high-fidelity music and professional-grade audio, capturing subtle nuances across a range of genres and intricate compositions | [Katie Nguyen](https:\u002F\u002Fgithub.com\u002Fkatiemn) | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fmodels\u002Flyria\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fcloud.google.com\u002Fvertex-ai\u002Fgenerative-ai\u002Fdocs\u002Fmusic\u002Fgenerate-music)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fsingularity\u002Fcomments\u002F1ku5tdw\u002Flyria_2_googles_new_ai_music_generator_sounds\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbiJJEbVjvko)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGoogleCloudPlatform\u002Fgenerative-ai\u002Fblob\u002Fmain\u002Faudio\u002Fmusic\u002Fgetting-started\u002Flyria2_music_generation.ipynb) | 06.11.2025 |\n| Vertex AI | Search brings together the power of deep information retrieval, state-of-the-art natural language processing, and the latest in LLM processing to understand user intent and return the most relevant results for the user | [Megha Agarwal](https:\u002F\u002Fgithub.com\u002Fagarwal22megha) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogleapis\u002Fpython-aiplatform?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogleapis\u002Fpython-aiplatform) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.16833)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fblog.google\u002Fproducts\u002Fsearch\u002Fimproving-search-next-20-years\u002Fv)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Ftechnologies\u002Fgemini\u002Fpro\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fcloud.google.com\u002Fgenerative-ai-app-builder\u002Fdocs\u002Fenterprise-search-introduction)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgoogle-cloud-aiplatform\u002F), [\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgoogle-cloud-discoveryengine\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLIivdWyY5sqJ1YuMdGjRwJ3fFYZ_vWQ62), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLIivdWyY5sqJAyUJbbsc8ZyGLNT4isnuB), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FHD_xreaLKb4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-3Olw-C4FN4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FXVO3zsHdvio)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGoogleCloudPlatform\u002Fgenerative-ai\u002Fblob\u002Fmain\u002Fsearch\u002Fvertexai-search-options\u002Fvertexai_search_options.ipynb) | 06.11.2025 |\n| ADK | Collection provides ready-to-use agents built on top of the Agent Development Kit, designed to accelerate your development process | \u003Cul>\u003Cli>[Equious](https:\u002F\u002Fgithub.com\u002FEquious)\u003C\u002Fli> \u003Cli>[Ankur Sharma](https:\u002F\u002Fgithub.com\u002Fankursharmas)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fadk-samples?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fadk-samples) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fdevelopers.googleblog.com\u002Fen\u002Fagent-development-kit-easy-to-build-multi-agent-applications\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fgoogle.github.io\u002Fadk-docs\u002F), [\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fdocs.cloud.google.com\u002Fagent-builder\u002Fagent-development-kit\u002Foverview)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fadk-go), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fadk-java)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@d3xvn\u002Fexploring-googles-agent-development-kit-adk-71a27a609920)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgoogle-adk\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fagentdevelopmentkit\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLOU2XLYxmsIIAPgM8FmtEcFTXLLzmh4DK), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FP4VFL9nIaIA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F44C8u0CDtSo), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FG9wnpfW6lZY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8rlNdKywldQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBiP4tKZKTvU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fcz2pKLPw994), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGeo8LzCHoMQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fadk-samples\u002Fblob\u002Fmain\u002Fpython\u002Fnotebooks\u002Fevaluation\u002Fuser_simulation_in_adk_evals.ipynb) | 05.11.2025 |\n| TRL | Set of tools to train trans","[![访问量](https:\u002F\u002Fhits.seeyoufarm.com\u002Fapi\u002Fcount\u002Fincr\u002Fbadge.svg?url=https:\u002F\u002Fgithub.com\u002Famrzv\u002Fawesome-colab-notebooks)](https:\u002F\u002Fhits.seeyoufarm.com)\n![awesome-colab-notebooks](https:\u002F\u002Fcount.getloli.com\u002Fget\u002F@awesome-colab-notebooks?theme=rule34)\n\n[![词云](images\u002Fcloud.svg)](images\u002Fcloud.svg)\n\n页面可能无法正常显示。请直接打开 [README.md](https:\u002F\u002Fgithub.com\u002Famrzv\u002Fawesome-colab-notebooks\u002Fblob\u002Fmain\u002FREADME.md) 文件\n# 用于机器学习实验的超赞 Colab 笔记本合集\n\n## 热门\n| 仓库 | 论文 | 软件包 |\n|---|---|---|\n| \u003Cul>\u003Cli>agent-starter-pack\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FGoogleCloudPlatform\u002Fagent-starter-pack?style=social)](https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fagent-starter-pack)\u003C\u002Fli> \u003Cli>PaddleHub\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPaddlePaddle\u002FPaddleHub?style=social)](https:\u002F\u002Fgithub.com\u002FPaddlePaddle\u002FPaddleHub)\u003C\u002Fli> \u003Cli>ritm-interactive-segmentation\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsupervisely-ecosystem\u002Fritm-interactive-segmentation?style=social)](https:\u002F\u002Fgithub.com\u002Fsupervisely-ecosystem\u002Fritm-interactive-segmentation)\u003C\u002Fli> \u003Cli>verl\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fvolcengine\u002Fverl?style=social)](https:\u002F\u002Fgithub.com\u002Fvolcengine\u002Fverl)\u003C\u002Fli> \u003Cli>Qwen3-Omni\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FQwenLM\u002FQwen3-Omni?style=social)](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen3-Omni)\u003C\u002Fli> \u003Cli>dinov3\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fdinov3?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdinov3)\u003C\u002Fli> \u003Cli>wmar\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fwmar?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fwmar)\u003C\u002Fli> \u003Cli>cwm\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fcwm?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fcwm)\u003C\u002Fli> \u003Cli>circle-guard-bench\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwhitecircle-ai\u002Fcircle-guard-bench?style=social)](https:\u002F\u002Fgithub.com\u002Fwhitecircle-ai\u002Fcircle-guard-bench)\u003C\u002Fli> \u003Cli>SAELens\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FjbloomAus\u002FSAELens?style=social)](https:\u002F\u002Fgithub.com\u002FjbloomAus\u002FSAELens)\u003C\u002Fli> \u003Cli>ARENA_3.0\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcallummcdougall\u002FARENA_3.0?style=social)](https:\u002F\u002Fgithub.com\u002Fcallummcdougall\u002FARENA_3.0)\u003C\u002Fli> \u003Cli>TabPFN\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fautoml\u002FTabPFN?style=social)](https:\u002F\u002Fgithub.com\u002Fautoml\u002FTabPFN)\u003C\u002Fli> \u003Cli>vjepa2\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fvjepa2?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fvjepa2)\u003C\u002Fli> \u003Cli>opik\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcomet-ml\u002Fopik?style=social)](https:\u002F\u002Fgithub.com\u002Fcomet-ml\u002Fopik)\u003C\u002Fli> \u003Cli>alphaevolve_results\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Falphaevolve_results?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Falphaevolve_results)\u003C\u002Fli> \u003Cli>SwanLab\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FSwanHubX\u002FSwanLab?style=social)](https:\u002F\u002Fgithub.com\u002FSwanHubX\u002FSwanLab)\u003C\u002Fli> \u003Cli>homework_fall2023\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fberkeleydeeprlcourse\u002Fhomework_fall2023?style=social)](https:\u002F\u002Fgithub.com\u002Fberkeleydeeprlcourse\u002Fhomework_fall2023)\u003C\u002Fli> \u003Cli>langgraph\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Flangchain-ai\u002Flanggraph?style=social)](https:\u002F\u002Fgithub.com\u002Flangchain-ai\u002Flanggraph)\u003C\u002Fli> \u003Cli>sglang\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsgl-project\u002Fsglang?style=social)](https:\u002F\u002Fgithub.com\u002Fsgl-project\u002Fsglang)\u003C\u002Fli> \u003Cli>part_1_ml_cv\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FDeepLearningSchool\u002Fpart_1_ml_cv?style=social)](https:\u002F\u002Fgithub.com\u002FDeepLearningSchool\u002Fpart_1_ml_cv)\u003C\u002Fli> \u003Cli>felix\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsimonmalberg\u002Ffelix?style=social)](https:\u002F\u002Fgithub.com\u002Fsimonmalberg\u002Ffelix)\u003C\u002Fli> \u003Cli>prompt-eng-interactive-tutorial\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fanthropics\u002Fprompt-eng-interactive-tutorial?style=social)](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fprompt-eng-interactive-tutorial)\u003C\u002Fli> \u003Cli>presidio\t[![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002Fpresidio?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fpresidio)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>trlX\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0ece95190254.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002F2023.emnlp-main.530)\u003C\u002Fli> \u003Cli>LLaMA Factory\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0b29ad8610c4.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002F2024.acl-demos.38)\u003C\u002Fli> \u003Cli>AWQ\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_260e9c4ff296.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3714983.3714987)\u003C\u002Fli> \u003Cli>EAT\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2fe6fd6c5a7e.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.02069)\u003C\u002Fli> \u003Cli>Fine-tuning a BERT\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1d8bb9ef5451.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002FN19-1423)\u003C\u002Fli> \u003Cli>Classify text with BERT\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1d8bb9ef5451.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002FN19-1423)\u003C\u002Fli> \u003Cli>Panini-Net\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_dc79c78aa8d7.png)](https:\u002F\u002Fdoi.org\u002F10.1609\u002Faaai.v36i3.20159)\u003C\u002Fli> \u003Cli>Gaussian Splatting\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0c34c9f6bf85.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3592433)\u003C\u002Fli> \u003Cli>GraphCast\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_95a3b8c45abb.png)](https:\u002F\u002Fdoi.org\u002F10.1126\u002Fscience.adi2336)\u003C\u002Fli> \u003Cli>BiRefNet\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_b8af1a40a030.png)](https:\u002F\u002Fdoi.org\u002F10.26599\u002FAIR.2024.9150038)\u003C\u002Fli> \u003Cli>SAHI\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_09c209fb3be9.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICIP46576.2022.9897990)\u003C\u002Fli> \u003Cli>DifFace\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_28a1a4fb4867.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTPAMI.2024.3432651)\u003C\u002Fli> \u003Cli>py-irt\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9acfca034502.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002F2021.acl-long.346)\u003C\u002Fli> \u003Cli>FreeInit\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cba6c3f73678.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-72646-0_22)\u003C\u002Fli> \u003Cli>NeRViS\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fd196cdcdbc9.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00230)\u003C\u002Fli> \u003Cli>LIDA\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_14e441d2aa9f.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002F2023.acl-demo.11)\u003C\u002Fli> \u003Cli>AST\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_bb66827c48fd.png)](https:\u002F\u002Fdoi.org\u002F10.21437\u002FInterspeech.2021-698)\u003C\u002Fli> \u003Cli>Geometry-Free View Synthesis\t[!](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9f50cfdb8dc3.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.01409)\u003C\u002Fli> \u003Cli>AudioLM\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_30e549055623.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTASLP.2023.3288409)\u003C\u002Fli> \u003Cli>deep-significance\t[![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fd26b66278d2.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002Fp19-1266)\u003C\u002Fli> \u003Cli>VRT\t[!](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_aedd82489829.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTIP.2024.3372454)\u003C\u002Fli> \u003Cli>GLIP\t[!](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_085ff51929b8.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01069)\u003C\u002Fli> \u003Cli>OWL-ViT\t[!](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9d15beeb274f.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20080-9_42)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>sam3\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fsam3?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fsam3\u002F)\u003C\u002Fli> \u003Cli>xminigrid\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fxminigrid?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fxminigrid\u002F)\u003C\u002Fli> \u003Cli>sglang\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fsglang?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fsglang\u002F)\u003C\u002Fli> \u003Cli>composer\t[![](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fcomposer?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fcomposer\u002F)\u003C\u002Fli> \u003Cli>a2a-sdk\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fa2a-sdk?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fa2a-sdk\u002F)\u003C\u002Fli> \u003Cli>dopamine-rl\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fdopamine-rl?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fdopamine-rl\u002F)\u003C\u002Fli> \u003Cli>google-cloud-discoveryengine\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fgoogle-cloud-discoveryengine?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fgoogle-cloud-discoveryengine\u002F)\u003C\u002Fli> \u003Cli>imagededup\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fimagededup?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fimagededup\u002F)\u003C\u002Fli> \u003Cli>xmanager\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fxmanager?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fxmanager\u002F)\u003C\u002Fli> \u003Cli>agent-starter-pack\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fagent-starter-pack?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fagent-starter-pack\u002F)\u003C\u002Fli> \u003Cli>torchgeo\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Ftorchgeo?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Ftorchgeo\u002F)\u003C\u002Fli> \u003Cli>rl-games\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Frl-games?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Frl-games\u002F)\u003C\u002Fli> \u003Cli>swanlab\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fswanlab?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fswanlab\u002F)\u003C\u002Fli> \u003Cli>transformer-lens\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Ftransformer-lens?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Ftransformer-lens\u002F)\u003C\u002Fli> \u003Cli>scikit-video\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fscikit-video?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fscikit-video\u002F)\u003C\u002Fli> \u003Cli>google-genai\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fgoogle-genai?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fgoogle-genai\u002F)\u003C\u002Fli> \u003Cli>tensorflow-graphics\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Ftensorflow-graphics?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Ftensorflow-graphics\u002F)\u003C\u002Fli> \u003Cli>neural-tangents\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fneural-tangents?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fneural-tangents\u002F)\u003C\u002Fli> \u003Cli>tensorflow-gnn\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Ftensorflow-gnn?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Ftensorflow-gnn\u002F)\u003C\u002Fli> \u003Cli>opik\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fopik?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fopik\u002F)\u003C\u002Fli> \u003Cli>sae-lens\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fsae-lens?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fsae-lens\u002F)\u003C\u002Fli> \u003Cli>csbdeep\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Fcsbdeep?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Fcsbdeep\u002F)\u003C\u002Fli> \u003Cli>langgraph\t[!](https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fdw\u002Flanggraph?style=flat&logo=pypi&label=%E2%80%8D&labelColor=f7f7f4&color=006dad)](https:\u002F\u002Fpypi.org\u002Flanggraph\u002F)\u003C\u002Fli>\u003C\u002Ful> |\n\n## 课程\n\u003Cdetails>\n\u003Csummary>课程\u003C\u002Fsummary>\n\n| name | description | authors | links | colaboratory | update |\n|------|-------------|:--------|:------|:------------:|:------:|\n| LLM Engineering Essentials course | 12-week course, created by experts from academia and industry, is designed specifically for developers and engineers | \u003Cul>\u003Cli>[Stanislav Fedotov](https:\u002F\u002Fgithub.com\u002Fst-fedotov)\u003C\u002Fli> \u003Cli>[Alexey Bukhtiyarov](https:\u002F\u002Fgithub.com\u002Fleshanbog)\u003C\u002Fli> \u003Cli>[Nikita Pavlichenko](https:\u002F\u002Fwww.pavlichenko.info\u002F)\u003C\u002Fli> \u003Cli>[Sergei Petrov](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsergei-petrov-12589210b\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Sergei Skvortsov](https:\u002F\u002Fgithub.com\u002Fsouthfreebird)\u003C\u002Fli> \u003Cli>[Alex Umnov](https:\u002F\u002Fgithub.com\u002FAlexUmnov)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNebius-Academy\u002FLLM-Engineering-Essentials?style=social)](https:\u002F\u002Fgithub.com\u002FNebius-Academy\u002FLLM-Engineering-Essentials) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002Fkarpathy\u002Fstatus\u002F1886192184808149383)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Facademy.nebius.com\u002Fllm-engineering-essentials)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FNebius-Academy\u002FLLM-Engineering-Essentials\u002Fblob\u002Fmain\u002Ftopic1\u002F1.1_intro_to_llm_apis.ipynb) | 22.05.2025 |\n| Understanding Language Models | Course deals with language models, in particular (but not exclusively so) on transformer-based language models like GPT-x or LLama | \u003Cul>\u003Cli>[Michael Franke](https:\u002F\u002Fmichael-franke.github.io\u002F)\u003C\u002Fli> \u003Cli>[Carsten Eickhoff](https:\u002F\u002Fgithub.com\u002Feickhoff)\u003C\u002Fli> \u003Cli>[Polina Tsvilodub](https:\u002F\u002Fpolina-tsvilodub.github.io\u002Fhome\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCogSciPrag\u002FUnderstanding-LLMs-course?style=social)](https:\u002F\u002Fgithub.com\u002FCogSciPrag\u002FUnderstanding-LLMs-course) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1301.3781), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2001.08361), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.18223)\u003C\u002Fli>\u003Cli>[book](https:\u002F\u002Fweb.stanford.edu\u002F~jurafsky\u002Fslp3\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fcogsciprag.github.io\u002FUnderstanding-LLMs-course\u002Fintro.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FzjkBMFhNj_g)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FCogSciPrag\u002FUnderstanding-LLMs-course\u002Fblob\u002Fmain\u002Funderstanding-llms\u002Ftutorials\u002F01-introduction.ipynb) | 17.04.2025 |\n| Practical RL | An open course on reinforcement learning in the wild | \u003Cul>\u003Cli>[Pavel Shvechikov](https:\u002F\u002Fgithub.com\u002Fpshvechikov)\u003C\u002Fli> \u003Cli>[Nikita Putintsev](https:\u002F\u002Fgithub.com\u002Fqwasser)\u003C\u002Fli> \u003Cli>[Alexander Fritsler](https:\u002F\u002Fgithub.com\u002FFritz449)\u003C\u002Fli> \u003Cli>[Oleg Vasilev](https:\u002F\u002Fme.svin.in\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dmitry Nikulin](https:\u002F\u002Fgithub.com\u002Fpastafarianist)\u003C\u002Fli> \u003Cli>[Mikhail Konobeev](https:\u002F\u002Fgithub.com\u002Fmknbv)\u003C\u002Fli> \u003Cli>[Ivan Kharitonov](https:\u002F\u002Fkharitonov-ivan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ravil Khisamov](https:\u002F\u002Fgithub.com\u002Fzshrav)\u003C\u002Fli> \u003Cli>[Anna Klepova](https:\u002F\u002Fgithub.com\u002Fq0o0p)\u003C\u002Fli> \u003Cli>[Fedor Ratnikov](https:\u002F\u002Fgithub.com\u002Fjustheuristic)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyandexdataschool\u002FPractical_RL?style=social)](https:\u002F\u002Fgithub.com\u002Fyandexdataschool\u002FPractical_RL) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1703.03864)\u003C\u002Fli>\u003Cli>[feedback](https:\u002F\u002Fdocs.google.com\u002Fforms\u002Fd\u002Fe\u002F1FAIpQLSdurWw97Sm9xCyYwC8g3iB5EibITnoPJW2IkOVQYE_kcXPh6Q\u002Fviewform)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFarama-Foundation\u002FGymnasium)\u003C\u002Fli>\u003Cli>[slides](https:\u002F\u002Fyadi.sk\u002Fd\u002FloPpY45J3EAYfU)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2pWv7GOvuf0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FaUrX-rP_ss4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FlfHX2hHRMVQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FejxfTy4lI6I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FzwYV11a__HQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FD58nLNLkb0I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FxpyKmjJuqhk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FKWlFFtmxjeg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FASQJTNAYdcA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyandexdataschool\u002FPractical_RL\u002Fblob\u002Fmaster\u002Fweek01_intro\u002Fdeep_crossentropy_method.ipynb) | 02.03.2025 |\n| Deep Learning School course (ML + CV) |  | [Nina Konovalova](https:\u002F\u002Fgithub.com\u002FNina-Konovalova) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FDeepLearningSchool\u002Fpart_1_ml_cv?style=social)](https:\u002F\u002Fgithub.com\u002FDeepLearningSchool\u002Fpart_1_ml_cv) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fdatasets\u002Fabrambeyer\u002Fopenintro-possum)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fdls.samcs.ru\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL0Ks75aof3TiHbkJ95vxNlQefujrj1N2w), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL5-da3qGB5ICCsgW1MxlZ0Hq8LL5U3u9y)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FDeepLearningSchool\u002Fpart_1_ml_cv\u002Fblob\u002Fmain\u002Fweek_01_ml_intro\u002FPractice\u002Fpart_1_data_analysis.ipynb) | 14.02.2025 |\n| Introduction to Deep Learning course |  | [Tatiana Gaintseva](https:\u002F\u002Fatmyre.github.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Faifoundation-edu\u002FDL_intro?style=social)](https:\u002F\u002Fgithub.com\u002Faifoundation-edu\u002FDL_intro) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@zxr.nju\u002Fwhat-is-the-kernel-trick-why-is-it-important-98a98db0961d), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fanalytics-vidhya\u002Fhow-relu-works-f317a947bdc6)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fplayground.tensorflow.org\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FKernel_method)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL0Ks75aof3ThuLLtLIVl_KPUDDQlTDyJI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Foa3IVkENrdM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FqvauAflpnyo), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQ7vT0--5VII)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Faifoundation-edu\u002FDL_intro\u002Fblob\u002Fmain\u002F01_Intro_to_Neural_Networks\u002F01_NN_intro.ipynb) | 24.01.2025 |\n| ARENA | Provide talented individuals with the skills, tools, and environment necessary for upskilling in ML engineering, for the purpose of contributing directly to AI alignment in technical roles | [Callum McDougall](https:\u002F\u002Fwww.perfectlynormal.co.uk\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcallummcdougall\u002FARENA_3.0?style=social)](https:\u002F\u002Fgithub.com\u002Fcallummcdougall\u002FARENA_3.0) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.00593)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fslack.svg\" alt=\"slack\" height=20\u002F>](https:\u002F\u002Fjoin.slack.com\u002Ft\u002Farena-uk\u002Fshared_invite\u002Fzt-2noug8mpy-TRYbCnc3pzj7ITNrZIjKww)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Farena-resources.notion.site\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1vuQOB2Gd7OcfzH2y9djXm9OdZA_DcxYz) | 30.12.2024 |\n| Deep Learning Course at the University of Amsterdam | Series of Jupyter notebooks that are designed to help you understanding the \"theory\" from the lectures by seeing corresponding implementations | \u003Cul>\u003Cli>[Pascal Mettes](https:\u002F\u002Fstaff.fnwi.uva.nl\u002Fp.s.m.mettes\u002F)\u003C\u002Fli> \u003Cli>[Melika Davood Zadeh](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FMelika-Davood-Zadeh)\u003C\u002Fli> \u003Cli>[Mohammadreza Salehidehnavi](https:\u002F\u002Fsmsd75.github.io\u002F)\u003C\u002Fli> \u003Cli>[Danilo de Goede](https:\u002F\u002Fivi.fnwi.uva.nl\u002Fvislab\u002Fauthor\u002Fdanilo-de-goede\u002F)\u003C\u002Fli> \u003Cli>[Phillip Lippe](https:\u002F\u002Fphlippe.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fphlippe\u002Fuvadlc_notebooks?style=social)](https:\u002F\u002Fgithub.com\u002Fphlippe\u002Fuvadlc_notebooks) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fuvadlc-notebooks.readthedocs.io\u002Fen\u002Flatest\u002Findex.html)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fuvadlc.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLdlPlO1QhMiAkedeu0aJixfkknLRxk1nA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fphlippe\u002Fuvadlc_notebooks\u002Fblob\u002Fmaster\u002Fdocs\u002Ftutorial_notebooks\u002Ftutorial2\u002FIntroduction_to_PyTorch.ipynb) | 17.10.2024 |\n| The Autodiff Cookbook | You'll go through a whole bunch of neat autodiff ideas that you can cherry pick for your own work, starting with the basics | \u003Cul>\u003Cli>[Alex Wiltschko](https:\u002F\u002Fgithub.com\u002Falexbw)\u003C\u002Fli> \u003Cli>[Matthew Johnson](http:\u002F\u002Fpeople.csail.mit.edu\u002Fmattjj\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fjax?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fjax\u002Fissues\u002F446#issuecomment-467105048) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1406.2572), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.04454), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.03451), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.07062)\u003C\u002Fli>\u003Cli>[book](https:\u002F\u002Fmitpress.mit.edu\u002Fsites\u002Fdefault\u002Ffiles\u002Ftitles\u002Fcontent\u002Fsicm_edition_2\u002Fbook.html), [book](https:\u002F\u002Fmitpress.mit.edu\u002Fbooks\u002Ffunctional-differential-geometry)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fjax#auto-vectorization-with-vmap), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhips\u002Fautograd)\u003C\u002Fli>\u003Cli>[tutorial](http:\u002F\u002Fvideolectures.net\u002Fdeeplearning2017_johnson_automatic_differentiation\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FTruncated_Newton_method), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FPullback_(differential_geometry)), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FHolomorphic_function), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FCauchy%E2%80%93Riemann_equations)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fjax\u002Fblob\u002Fmain\u002Fdocs\u002Fnotebooks\u002Fautodiff_cookbook.ipynb) | 20.09.2024 |\n| Machine Learning Simplified | A Gentle Introduction to Supervised Learning | [Andrew Wolf](https:\u002F\u002F5x12.ai\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002F5x12\u002Fthemlsbook?style=social)](https:\u002F\u002Fgithub.com\u002F5x12\u002Fthemlsbook) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fgeekculture\u002Fi-found-a-great-machine-learning-book-deed11db2688)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FPython\u002Fcomments\u002Ft8st9l\u002Fi_wrote_a_book_on_machine_learning_w_python_code\u002F), [\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Flearnmachinelearning\u002Fcomments\u002Fsnxlly\u002Fmachine_learning_simplified_book\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fwww.themlsbook.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002F5x12\u002Fthemlsbook\u002Fblob\u002Fmaster\u002Fchapter2\u002Fknn.ipynb) | 29.08.2024 |\n| Anthropic courses | Anthropic's educational courses | [Anthropic](https:\u002F\u002Fwww.anthropic.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fanthropics\u002Fcourses?style=social)](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fcourses) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fdocs.anthropic.com\u002Fen\u002Fdocs\u002Fresources\u002Fcourses)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fanthropic-sdk-python)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fanthropic\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FClaudeAI\u002Fcomments\u002F1f7czsx\u002Fanthropics_official_educational_courses_on_prompt\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fanthropics\u002Fcourses\u002Fblob\u002Fmaster\u002Fanthropic_api_fundamentals\u002F01_getting_started.ipynb) | 22.08.2024 |\n| mlcourse.ai | Open Machine Learning Course | [Yury Kashnitsky](https:\u002F\u002Fyorko.github.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYorko\u002Fmlcourse.ai?style=social)](https:\u002F\u002Fgithub.com\u002FYorko\u002Fmlcourse.ai) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fhabr.com\u002Fcompany\u002Fods\u002Fblog\u002F344044\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fkashnitsky\u002Fmlcourse)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fopen-machine-learning-course)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmlcourse.ai\u002Fbook\u002Findex.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fslack.svg\" alt=\"slack\" height=20\u002F>](https:\u002F\u002Fopendatascience.slack.com\u002Farchives\u002FC91N8TL83\u002Fp1567408586359500)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLVlY_7IJCMJeRfZ68eVfEcu-UcN9BbwiX)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FYorko\u002Fmlcourse.ai\u002Fblob\u002Fmain\u002Fjupyter_english\u002Ftopic01_pandas_data_analysis\u002Ftopic1_pandas_data_analysis.ipynb) | 19.08.2024 |\n| Deep RL Course | The Hugging Face Deep Reinforcement Learning Course | \u003Cul>\u003Cli>[Thomas Simonini](https:\u002F\u002Fwww.simoninithomas.com\u002F)\u003C\u002Fli> \u003Cli>[Omar Sanseviero](https:\u002F\u002Fosanseviero.github.io\u002Fhackerllama\u002F)\u003C\u002Fli> \u003Cli>[Sayak Paul](https:\u002F\u002Fsayak.dev\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhuggingface\u002Fdeep-rl-class?style=social)](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fdeep-rl-class) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FydHrjt3WP5)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Falex-petrenko\u002Fsample-factory)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdeep-rl-course\u002Funit0\u002Fintroduction), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fhuggingface-projects\u002FDeep-Reinforcement-Learning-Leaderboard)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fpytorch.org\u002Ftutorials\u002Fbeginner\u002Fdeep_learning_60min_blitz.html)\u003C\u002Fli>\u003Cli>[syllabus](https:\u002F\u002Fsimoninithomas.github.io\u002Fdeep-rl-course)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2GwBez0D20A), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FCsuIANBnSq8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAQKAOXJa6qg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fhuggingface\u002Fdeep-rl-class\u002Fblob\u002Fmain\u002Fnotebooks\u002Funit1\u002Funit1.ipynb) | 24.06.2024 |\n| Anthropic's Prompt Engineering Interactive Tutorial | Course is intended to provide you with a comprehensive step-by-step understanding of how to engineer optimal prompts within Claude | [Anthropic](https:\u002F\u002Fwww.anthropic.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fanthropics\u002Fprompt-eng-interactive-tutorial?style=social)](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fprompt-eng-interactive-tutorial) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fdocs.anthropic.com\u002Fen\u002Fdocs\u002Fbuild-with-claude\u002Fprompt-engineering\u002Foverview)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Fanthropic-sdk-python)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fcoding-nexus\u002Fmastering-prompt-engineering-for-ai-a-free-hands-on-guide-with-anthropics-tutorial-6f2cabcbde5a)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FT9aRN5JkmL8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhkhDdcM5V94)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fanthropics\u002Fprompt-eng-interactive-tutorial\u002Fblob\u002Fmaster\u002FAnthropic%201P\u002F00_Tutorial_How-To.ipynb) | 02.04.2024 |\n| Generative AI for Beginners - A Course | A 12 Lesson course teaching everything you need to know to start building Generative AI applications | [microsoft](https:\u002F\u002Fwww.microsoft.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002Fxr-development-for-beginners?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fxr-development-for-beginners) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Faka.ms\u002Fgenai-discord)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FWeb-Dev-For-Beginners)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmicrosoft.github.io\u002Fgenerative-ai-for-beginners\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmicrosoft\u002Fgenerative-ai-for-beginners\u002Fblob\u002Fmain\u002F06-text-generation-apps\u002Fnotebook-azure-openai.ipynb) | 22.02.2024 |\n| Deep Reinforcement Learning | CS 285 at UC Berkeley | \u003Cul>\u003Cli>[Sergey Levine](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~svlevine\u002F)\u003C\u002Fli> \u003Cli>[Kyle Stachowicz](https:\u002F\u002Fkylestaschowicz.com)\u003C\u002Fli> \u003Cli>[Vivek Myers](https:\u002F\u002Fvmyers.github.io)\u003C\u002Fli> \u003Cli>[Joey Hong](https:\u002F\u002Fjoeyhong123.github.io)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Kevin Black](https:\u002F\u002Fkvablack.github.io)\u003C\u002Fli> \u003Cli>[Michael Janner](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~janner\u002F)\u003C\u002Fli> \u003Cli>[Vitchyr Pong](https:\u002F\u002Fvitchyr.github.io)\u003C\u002Fli> \u003Cli>[Aviral Kumar](https:\u002F\u002Faviralkumar2907.github.io)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fberkeleydeeprlcourse\u002Fhomework_fall2023?style=social)](https:\u002F\u002Fgithub.com\u002Fberkeleydeeprlcourse\u002Fhomework_fall2023) \u003Cul>\u003Cli>[coursera](https:\u002F\u002Fwww.coursera.org\u002Fcourse\u002Fneuralnets), [coursera](https:\u002F\u002Fwww.coursera.org\u002Flearn\u002Fmachine-learning\u002F)\u003C\u002Fli>\u003Cli>[website](http:\u002F\u002Frail.eecs.berkeley.edu\u002Fdeeprlcourse\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL_iWQOsE6TfVYGEGiAOMaOzzv41Jfm_Ps), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutube.com\u002Fplaylist?list=PL_iWQOsE6TfX7MaC6C3HcdOf1g337dlC9)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fberkeleydeeprlcourse\u002Fhomework_fall2023\u002Fblob\u002Fmaster\u002Fhw1\u002Fcs285\u002Fscripts\u002Frun_hw1.ipynb) | 29.08.2023 |\n| npNLG | The course introduces the basics of NLG, neural language models and their implementation in PyTorch, as well as a selection of recent pragmatic neural NLG approaches | [Michael Franke](https:\u002F\u002Fgithub.com\u002Fmichael-franke) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a1b5a7212442.png)](https:\u002F\u002Fdoi.org\u002F10.1126\u002Fscience.1218633) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmichael-franke\u002FnpNLG?style=social)](https:\u002F\u002Fgithub.com\u002Fmichael-franke\u002FnpNLG) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.09867)\u003C\u002Fli>\u003Cli>[book](https:\u002F\u002Fmichael-franke.github.io\u002FnpNLG\u002F000-intro.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmichael-franke\u002FnpNLG\u002Fblob\u002Fmain\u002Fneural_pragmatic_nlg\u002F01-prob-prag\u002F01b-RSA-vanilla.ipynb) | 09.11.2022 |\n| DSP theory | Theory of digital signal processing: signals, filtration (IIR, FIR, CIC, MAF), transforms (FFT, DFT, Hilbert, Z-transform) etc | \u003Cul>\u003Cli>[Alexander Kapitanov](https:\u002F\u002Fgithub.com\u002Fhukenovs)\u003C\u002Fli> \u003Cli>[Vladimir Fadeev](https:\u002F\u002Fgithub.com\u002Fkirlf)\u003C\u002Fli> \u003Cli>[Karina Kvanchiani](https:\u002F\u002Fgithub.com\u002Fkarinakvanchiani)\u003C\u002Fli> \u003Cli>[Elizaveta Petrova](https:\u002F\u002Fgithub.com\u002Fkleinsbotle)\u003C\u002Fli> \u003Cli>[Andrei Makhliarchuk](https:\u002F\u002Fgithub.com\u002Fanotherhelloworld)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhukenovs\u002Fdsp-theory?style=social)](https:\u002F\u002Fgithub.com\u002Fhukenovs\u002Fdsp-theory) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fhabr.com\u002Fru\u002Farticles\u002F460445\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fhukenovs\u002Fdsp-theory\u002Fblob\u002Fmaster\u002Fsrc\u002Fdsp_theory_1_signals.ipynb) | 18.10.2022 |\n| Machine learning course | This course is broad and shallow, but author will provide additional links so that you can deepen your understanding of the ML method you need | [Тимчишин Віталій](https:\u002F\u002Fgithub.com\u002Ffbeilstein) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffbeilstein\u002Fmachine_learning?style=social)](https:\u002F\u002Fgithub.com\u002Ffbeilstein\u002Fmachine_learning) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fvas3k.com\u002Fblog\u002Fmachine_learning\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLkDeTjsoxDVgnb2lIYo9-1l4XYhrIyS6A), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-RdOwhmqP5s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FR13BD8qKeTg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FZkjP5RJLQF4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJ4Wdy0Wc_xQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FmBcLRGuAFUk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYIGtalP1mv0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYz5pySyEtsU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fx5zLaWT5KPs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyBwpo-L80Mc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL3FW7Lu3i5JvHM8ljYj-zLfQRF3EO8sYv)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffbeilstein\u002Fmachine_learning\u002Fblob\u002Fmaster\u002Flecture_01_introduction.ipynb) | 02.09.2021 |\n| Udacity Deep Learning class with TensorFlow | Learn how to apply deep learning to solve complex problems | [Mark Daoust](https:\u002F\u002Fgithub.com\u002FMarkDaoust) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Fexamples?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fexamples\u002Ftree\u002Fmaster\u002Fcourses\u002Fudacity_deep_learning) \u003Cul>\u003Cli>[data](http:\u002F\u002Fyaroslavvb.blogspot.com\u002F2011\u002F09\u002Fnotmnist-dataset.html), [data](http:\u002F\u002Fyann.lecun.com\u002Fexdb\u002Fmnist\u002F)\u003C\u002Fli>\u003Cli>[udacity](https:\u002F\u002Fwww.udacity.com\u002Fcourse\u002Fdeep-learning--ud730)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLAwxTw4SYaPn_OWPFT9ulXLuQrImzHfOV)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fexamples\u002Fblob\u002Fmaster\u002Fcourses\u002Fudacity_deep_learning\u002F1_notmnist.ipynb) | 20.01.2021 |\n| Intro to TensorFlow for Deep Learning | Dive into deep learning with this practical course on TensorFlow and the Keras API | \u003Cul>\u003Cli>[Magnus Hyttsten](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fmagnushyttsten)\u003C\u002Fli> \u003Cli>[Juan Delgado](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fjuan-delgado-845749179)\u003C\u002Fli> \u003Cli>[Paige Bailey](https:\u002F\u002Fwebpaige.dev\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Fexamples?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fexamples\u002Ftree\u002Fmaster\u002Fcourses\u002Fudacity_intro_to_tensorflow_for_deep_learning) \u003Cul>\u003Cli>[udacity](https:\u002F\u002Fwww.udacity.com\u002Fcourse\u002Fintro-to-tensorflow-for-deep-learning--ud187)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fexamples\u002Fblob\u002Fmaster\u002Fcourses\u002Fudacity_intro_to_tensorflow_for_deep_learning\u002Fl02c01_celsius_to_fahrenheit.ipynb) | 09.09.2020 |\n| Introduction to TensorFlow Lite | Learn how to deploy deep learning models on mobile and embedded devices with TensorFlow Lite | \u003Cul>\u003Cli>[Daniel Situnayake](https:\u002F\u002Fsitunayake.com\u002F)\u003C\u002Fli> \u003Cli>[Paige Bailey](https:\u002F\u002Fwebpaige.dev\u002F)\u003C\u002Fli> \u003Cli>[Juan Delgado](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fjuan-delgado-845749179)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Fexamples?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fexamples\u002Ftree\u002Fmaster\u002Fcourses\u002Fudacity_intro_to_tensorflow_lite) \u003Cul>\u003Cli>[udacity](https:\u002F\u002Fwww.udacity.com\u002Fcourse\u002Fintro-to-tensorflow-lite--ud190)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fexamples\u002Fblob\u002Fmaster\u002Fcourses\u002Fudacity_intro_to_tensorflow_lite\u002Ftflite_c01_linear_regression.ipynb) | 09.09.2020 |\n| NYU-DLSP20 | This course concerns the latest techniques in deep learning and representation learning, focusing on supervised and unsupervised deep learning, embedding methods, metric learning, convolutional and recurrent nets, with applications to computer vision, natural language understanding, and speech recognition | \u003Cul>\u003Cli>[Yann LeCun](https:\u002F\u002Fyann.lecun.com\u002F)\u003C\u002Fli> \u003Cli>[Alfredo Canziani](https:\u002F\u002Fatcold.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAtcold\u002FNYU-DLSP20?style=social)](https:\u002F\u002Fgithub.com\u002FAtcold\u002FNYU-DLSP20) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FCthuqsX8Pb)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAtcold\u002FNYU-DLSP21), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAtcold\u002FNYU-DLFL22)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FNYU_DeepLearning\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fatcold.github.io\u002FNYU-DLSP20\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLLHTzKZzVU9eaEyErdV26ikyolxOsz6mq)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAtcold\u002FNYU-DLSP20\u002Fblob\u002Fmaster\u002F00-logic_neuron_programming.ipynb) | 30.10.2019 |\n\n\u003C\u002Fdetails>\n\n\n\n## 研究\n\u003Cdetails>\n\u003Csummary>研究\u003C\u002Fsummary>\n\n| name | description | authors | links | colaboratory | update |\n|------|-------------|:--------|:------|:------------:|:------:|\n| GigaAM | SSL pretraining framework that leverages masked language modeling with targets derived from a speech recognition model | \u003Cul>\u003Cli>[Aleksandr Kutsakov](https:\u002F\u002Fgithub.com\u002FAlexander4127)\u003C\u002Fli> \u003Cli>[Alexandr Maximenko](https:\u002F\u002Fgithub.com\u002FAlexMaximenko)\u003C\u002Fli> \u003Cli>[Georgii Gospodinov](https:\u002F\u002Fgithub.com\u002Fgeorgygospodinov)\u003C\u002Fli> \u003Cli>[Pavel Bogomolov](https:\u002F\u002Fgithub.com\u002FBobrosoft98)\u003C\u002Fli> \u003Cli>[Fyodor Minkin](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FFyodor-Minkin)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsalute-developers\u002FGigaAM?style=social)](https:\u002F\u002Fgithub.com\u002Fsalute-developers\u002FGigaAM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.01192), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.08100), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1211.3711), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.11477), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.01192)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fhabr.com\u002Fru\u002Fcompanies\u002Fsberdevices\u002Farticles\u002F805569)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fai-sage\u002FGigaAM-v3), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fai-sage\u002FGigaAM-v3)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgigaam\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FO7NSH2SAwRc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FPvZuTUnZa2Q), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FktO4Mx6UMNk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F7NEvFNtwRTA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsalute-developers\u002FGigaAM\u002Fblob\u002Fmain\u002Fcolab_example.ipynb) | 20.11.2025 |\n| Segment Anything 3 | Unified model that detects, segments, and tracks objects in images and videos based on concept prompts, which we define as either short noun phrases (e.g., “yellow school bus”), image exemplars, or a combination of both | \u003Cul>\u003Cli>[Nicolas Carion](https:\u002F\u002Fwww.nicolascarion.com\u002F)\u003C\u002Fli> \u003Cli>[Laura Gustafson](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=c8IpF9gAAAAJ)\u003C\u002Fli> \u003Cli>[Yuan-Ting Hu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=E8DVVYQAAAAJ)\u003C\u002Fli> \u003Cli>[Shoubhik Debnath](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=fb6FOfsAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ronghang Hu](https:\u002F\u002Fronghanghu.com\u002F)\u003C\u002Fli> \u003Cli>[Didac Suris](https:\u002F\u002Fwww.didacsuris.com\u002F)\u003C\u002Fli> \u003Cli>[Chaitanya Ryali](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4LWx24UAAAAJ)\u003C\u002Fli> \u003Cli>[Kalyan Vasudev Alwala](https:\u002F\u002Fscholar.google.co.in\u002Fcitations?user=m34oaWEAAAAJ)\u003C\u002Fli> \u003Cli>[Haitham Khedr](https:\u002F\u002Fhkhedr.com\u002F)\u003C\u002Fli> \u003Cli>[Andrew Huang]()\u003C\u002Fli> \u003Cli>[Jie Lei](https:\u002F\u002Fjayleicn.github.io)\u003C\u002Fli> \u003Cli>[Tengyu Ma](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=VeTSl0wAAAAJ)\u003C\u002Fli> \u003Cli>[Baishan Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=BC5wDu8AAAAJ)\u003C\u002Fli> \u003Cli>[Arpit Kalla](https:\u002F\u002Fgithub.com\u002Farpitkalla)\u003C\u002Fli> \u003Cli>[Markus Marks](https:\u002F\u002Fdamaggu.github.io)\u003C\u002Fli> \u003Cli>[Joseph Greer](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=guL96CkAAAAJ)\u003C\u002Fli> \u003Cli>[Meng Wang]()\u003C\u002Fli> \u003Cli>[Peize Sun](https:\u002F\u002Fpeizesun.github.io)\u003C\u002Fli> \u003Cli>[Roman Rädle](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Tpt57v0AAAAJ)\u003C\u002Fli> \u003Cli>[Triantafyllos Afouras](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~afourast)\u003C\u002Fli> \u003Cli>[Effrosyni Mavroudi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vYRzGGEAAAAJ)\u003C\u002Fli> \u003Cli>[Katherine Xu](https:\u002F\u002Fk8xu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tsung-Han Wu](https:\u002F\u002Fpatrickthwu.com\u002F)\u003C\u002Fli> \u003Cli>[Yu Zhou](https:\u002F\u002Fyu-bryan-zhou.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liliane Momeni](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Lb-KgVYAAAAJ)\u003C\u002Fli> \u003Cli>[Rishi Hazra](https:\u002F\u002Frishihazra.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shuangrui Ding](https:\u002F\u002Fmark12ding.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sagar Vaze](https:\u002F\u002Fsgvaze.github.io\u002F)\u003C\u002Fli> \u003Cli>[Francois Porcher](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=LgHZ8hUAAAAJ)\u003C\u002Fli> \u003Cli>[Feng Li](https:\u002F\u002Ffengli-ust.github.io\u002F)\u003C\u002Fli> \u003Cli>[Siyuan Li](https:\u002F\u002Fsiyuanliii.github.io\u002F)\u003C\u002Fli> \u003Cli>[Aishwarya Kamath](https:\u002F\u002Fashkamath.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ho Kei Cheng](https:\u002F\u002Fhkchengrex.com\u002F)\u003C\u002Fli> \u003Cli>[Piotr Dollar](https:\u002F\u002Fpdollar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nikhila Ravi](https:\u002F\u002Fnikhilaravi.com\u002F)\u003C\u002Fli> \u003Cli>[Kate Saenko](https:\u002F\u002Fai.bu.edu\u002Fksaenko.html)\u003C\u002Fli> \u003Cli>[Pengchuan Zhang](https:\u002F\u002Fpzzhang.github.io\u002Fpzzhang\u002F)\u003C\u002Fli> \u003Cli>[Christoph Feichtenhofer](https:\u002F\u002Ffeichtenhofer.github.io)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fsam3?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fsam3) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2511.16719)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Faidemos.meta.com\u002Fsegment-anything)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002Fsam3), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Ffacebook\u002FSACo-Gold), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Ffacebook\u002FSACo-Silver), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Ffacebook\u002FSACo-VEval)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-sam-3-segment-anything-with-concepts-18e9501df00e)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresearch\u002Fpublications\u002Fsam-3-segment-anything-with-concepts\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fsam3\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fsegment-anything-model-3\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fabout.fb.com\u002Fnews\u002F2025\u002F11\u002Fnew-sam-models-detect-objects-create-3d-reconstructions\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fsam3\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002F1p1cfvx\u002Fr_segment_anything_model_3_sam_3_is_released\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLIsFyaUd-B30iOHlOHOb3OX6V_dum_DtN), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIATrWxEDpu4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fsam3\u002Fblob\u002Fmain\u002Fexamples\u002Fsam3_for_sam2_video_task_example.ipynb) | 19.11.2025 |\n| AlphaFold | Highly accurate protein structure prediction | \u003Cul>\u003Cli>[John Jumper](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=a5goOh8AAAAJ)\u003C\u002Fli> \u003Cli>[Richard Evans](http:\u002F\u002Fwww.doc.ic.ac.uk\u002F~re14\u002F)\u003C\u002Fli> \u003Cli>[Alexander Pritzel](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=GPgAyU0AAAAJ)\u003C\u002Fli> \u003Cli>[Tim Green](http:\u002F\u002Ftfgg.me\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Michael Figurnov](https:\u002F\u002Ffigurnov.ru\u002F)\u003C\u002Fli> \u003Cli>[Olaf Ronneberger](https:\u002F\u002Flmb.informatik.uni-freiburg.de\u002Fpeople\u002Fronneber\u002F)\u003C\u002Fli> \u003Cli>[Kathryn Tunyasuvunakool](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=eEqNGagAAAAJ)\u003C\u002Fli> \u003Cli>[Russ Bates](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Koes5ewAAAAJ)\u003C\u002Fli> \u003Cli>[Augustin Žídek](https:\u002F\u002Faugustin.zidek.eu\u002F)\u003C\u002Fli> \u003Cli>[Anna Potapenko](http:\u002F\u002Fapotapenko.com\u002F)\u003C\u002Fli> \u003Cli>[Alex Bridgland](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=VWmXKPMAAAAJ)\u003C\u002Fli> \u003Cli>[Clemens Meyer](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=EWLZiM8AAAAJ)\u003C\u002Fli> \u003Cli>[Simon Kohl](https:\u002F\u002Fwww.simonkohl.com\u002F)\u003C\u002Fli> \u003Cli>[Andrew Ballard](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=syjQhAMAAAAJ)\u003C\u002Fli> \u003Cli>[Bernardino Romera-Paredes](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fromeraparedes\u002F)\u003C\u002Fli> \u003Cli>[Stanislav Nikolov](https:\u002F\u002Fscholar.google.co.uk\u002Fcitations?user=O-b7pBEAAAAJ)\u003C\u002Fli> \u003Cli>[Rishub Jain](http:\u002F\u002Frishub.me\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_80dd955e0b2f.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41586-021-03819-2) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepmind\u002Falphafold?style=social)](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Falphafold) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fdeepmind.com\u002Fblog\u002Farticle\u002Falphafold-a-solution-to-a-50-year-old-grand-challenge-in-biology), [blog post](https:\u002F\u002Fdeepmind.com\u002Fblog\u002Farticle\u002Fputting-the-power-of-alphafold-into-the-worlds-hands)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsokrypton\u002FColabFold), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Ftree), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fchex)\u003C\u002Fli>\u003Cli>[paper](https:\u002F\u002Fwww.nature.com\u002Farticles\u002Fs41586-021-03828-1)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fmethod\u002Falphafold)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FAlphaFold)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=gg7WjuFs8F4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=B9PL__gVxLI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsokrypton\u002FColabFold\u002Fblob\u002Fmain\u002FAlphaFold2.ipynb) | 23.10.2025 |\n| OWL-ViT | Simple Open-Vocabulary Object Detection with Vision Transformers | \u003Cul>\u003Cli>[Matthias Minderer](http:\u002F\u002Fmatthias.minderer.net\u002F)\u003C\u002Fli> \u003Cli>[Alexey Gritsenko](https:\u002F\u002Fgithub.com\u002FAlexeyG)\u003C\u002Fli> \u003Cli>[Austin Stone](https:\u002F\u002Fgithub.com\u002FAustinCStone)\u003C\u002Fli> \u003Cli>[Maxim Neumann](https:\u002F\u002Fgithub.com\u002Fmaximneumann)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dirk Weissenborn](https:\u002F\u002Fgithub.com\u002Fdirkweissenborn)\u003C\u002Fli> \u003Cli>[Alexey Dosovitskiy](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=FXNJRDoAAAAJ)\u003C\u002Fli> \u003Cli>[Aravindh Mahendran](https:\u002F\u002Fgithub.com\u002Faravindhm)\u003C\u002Fli> \u003Cli>[Anurag Arnab](https:\u002F\u002Fgithub.com\u002Fanuragarnab)\u003C\u002Fli> \u003Cli>[Mostafa Dehghani](https:\u002F\u002Fmostafadehghani.com\u002F)\u003C\u002Fli> \u003Cli>[Zhuoran Shen](https:\u002F\u002Fcmsflash.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xiao Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ukyXqzMAAAAJ)\u003C\u002Fli> \u003Cli>[Xiaohua Zhai](https:\u002F\u002Fgithub.com\u002Fxiaohuazhai)\u003C\u002Fli> \u003Cli>[Thomas Kipf](https:\u002F\u002Ftkipf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Neil Houlsby](https:\u002F\u002Fneilhoulsby.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9d15beeb274f.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20080-9_42) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.06230)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Fowlvit)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fhuggingface\u002Fnotebooks\u002Fblob\u002Fmain\u002Fexamples\u002Fzeroshot_object_detection_with_owlvit.ipynb) | 29.09.2025 |\n| CWM | Code World Model, a 32-billion-parameter open-weights LLM, to advance research on code generation with world models | \u003Cul>\u003Cli>[Jade Copet](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=GRMLwjAAAAAJ)\u003C\u002Fli> \u003Cli>[Quentin Carbonneaux](https:\u002F\u002Ffr.linkedin.com\u002Fin\u002Fquentin-carbonneaux-36730717a)\u003C\u002Fli> \u003Cli>[Gal Cohen](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fgal-cohen-29b980122)\u003C\u002Fli> \u003Cli>[Jonas Gehring](https:\u002F\u002Fjgehring.net)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jacob Kahn](https:\u002F\u002Fjacobkahn.me)\u003C\u002Fli> \u003Cli>[Jannik Kossen](https:\u002F\u002Fopenreview.net\u002Fprofile?id=~Jannik_Kossen2)\u003C\u002Fli> \u003Cli>[Felix Kreuk](https:\u002F\u002Fopenreview.net\u002Fprofile?id=~Felix_Kreuk1)\u003C\u002Fli> \u003Cli>[Emily McMilin](https:\u002F\u002F2dot71mily.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michel Meyer](https:\u002F\u002Fwww.frenchweb.fr\u002Fa-la-rencontre-de-michel-meyer-technical-program-manager-pour-facebook-base-a-san-francisco\u002F427606)\u003C\u002Fli> \u003Cli>[Yuxiang Wei](https:\u002F\u002Fyuxiang.cs.illinois.edu)\u003C\u002Fli> \u003Cli>[David Zhang](https:\u002F\u002Fdavzha.netlify.app)\u003C\u002Fli> \u003Cli>[Kunhao Zheng](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=zDy4jSYAAAAJ)\u003C\u002Fli> \u003Cli>[Jordi Armengol Estape](http:\u002F\u002Fjordiae.com)\u003C\u002Fli> \u003Cli>[Pedram Bashiri](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=weIwEtgAAAAJ)\u003C\u002Fli> \u003Cli>[Maximilian Beck](https:\u002F\u002Fmaximilianmbeck.github.io)\u003C\u002Fli> \u003Cli>[Pierre Chambon](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5e_2WW4AAAAJ)\u003C\u002Fli> \u003Cli>[Abhishek Charnalia](https:\u002F\u002Fuk.linkedin.com\u002Fin\u002Fabhishekcharnalia)\u003C\u002Fli> \u003Cli>[Chris Cummins](https:\u002F\u002Fchriscummins.cc)\u003C\u002Fli> \u003Cli>[Juliette Decugis](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QXSA4p4AAAAJ)\u003C\u002Fli> \u003Cli>[Zacharias Fisches](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=nYfsAVcAAAAJ)\u003C\u002Fli> \u003Cli>[François Fleuret](https:\u002F\u002Ffleuret.org)\u003C\u002Fli> \u003Cli>[Fabian Gloeckle](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=_dF0fX4AAAAJ)\u003C\u002Fli> \u003Cli>[Alex Gu](https:\u002F\u002Fminimario.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michael Hassid](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5d93np0AAAAJ)\u003C\u002Fli> \u003Cli>[Daniel Haziza](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2eSKdFMAAAAJ)\u003C\u002Fli> \u003Cli>[Badr Youbi Idrissi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=xB5UivYAAAAJ)\u003C\u002Fli> \u003Cli>[Christian Keller](https:\u002F\u002Ffr.linkedin.com\u002Fin\u002Fkellerchristian)\u003C\u002Fli> \u003Cli>[Rahul Kindi](https:\u002F\u002Fgithub.com\u002Frkindi)\u003C\u002Fli> \u003Cli>[Hugh Leather](https:\u002F\u002Fhomepages.inf.ed.ac.uk\u002Fhleather\u002F)\u003C\u002Fli> \u003Cli>[Gallil Maimon](https:\u002F\u002Fpages.cs.huji.ac.il\u002Fgallilmaimon\u002F)\u003C\u002Fli> \u003Cli>[Aram Markosyan](https:\u002F\u002Fwww.markosyanaram.com\u002F)\u003C\u002Fli> \u003Cli>[Francisco Massa](https:\u002F\u002Ffmassa.github.io)\u003C\u002Fli> \u003Cli>[Pierre-Emmanuel Mazaré](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=IUod9t8AAAAJ)\u003C\u002Fli> \u003Cli>[Vegard Mella](https:\u002F\u002Fgithub.com\u002Ftscmoo)\u003C\u002Fli> \u003Cli>[Naila Murray](https:\u002F\u002Fnailamurray.github.io)\u003C\u002Fli> \u003Cli>[Keyur Muzumdar](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5mkv6-kAAAAJ)\u003C\u002Fli> \u003Cli>[Peter O’Hearn](https:\u002F\u002Fai.meta.com\u002Fpeople\u002F567943484717417\u002Fpeter-ohearn\u002F)\u003C\u002Fli> \u003Cli>[Matteo Pagliardini](https:\u002F\u002Fmatteopagliardini.github.io)\u003C\u002Fli> \u003Cli>[Dmitrii Pedchenko](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=izGGAnQAAAAJ)\u003C\u002Fli> \u003Cli>[Tal Remez](https:\u002F\u002Ftalremez.github.io\u002F)\u003C\u002Fli> \u003Cli>[Volker Seeker](https:\u002F\u002Fwww.volkerseeker.com\u002F)\u003C\u002Fli> \u003Cli>[Marco Selvi](https:\u002F\u002Fgithub.com\u002Fmarcoselvi)\u003C\u002Fli> \u003Cli>[Oren Sultan](https:\u002F\u002Fgithub.com\u002Forensul)\u003C\u002Fli> \u003Cli>[Sida Wang](https:\u002F\u002Fai.meta.com\u002Fpeople\u002F327339230340379\u002Fsida-wang\u002F)\u003C\u002Fli> \u003Cli>[Luca Wehrstedt](https:\u002F\u002Fgithub.com\u002Flw)\u003C\u002Fli> \u003Cli>[Ori Yoran](https:\u002F\u002Fwww.oriyoran.com\u002F)\u003C\u002Fli> \u003Cli>[Lingming Zhang](https:\u002F\u002Flingming.cs.illinois.edu)\u003C\u002Fli> \u003Cli>[Taco Cohen](https:\u002F\u002Ftacocohen.com)\u003C\u002Fli> \u003Cli>[Yossi Adi](https:\u002F\u002Fai.meta.com\u002Fpeople\u002F1930846650664668\u002Fyossi-adi\u002F)\u003C\u002Fli> \u003Cli>[Gabriel Synnaeve](https:\u002F\u002Fai.meta.com\u002Fpeople\u002F1447559096135307\u002Fgabriel-synnaeve\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fcwm?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fcwm) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002Fcwm), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002Fcwm-sft), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002Fcwm-pretrain)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science-in-your-pocket\u002Fmeta-code-world-models-released-f988a2f92e71), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fnoailabs.medium.com\u002Fcode-world-model-the-dawn-of-self-aware-software-b07a37cfd600)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresearch\u002Fpublications\u002Fcwm\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresources\u002Fmodels-and-libraries\u002Fcwm-downloads)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FAIatMeta\u002Fstatus\u002F1970963571753222319)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FvUYhfbj52E0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F9yRxXCJDkS0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fcwm\u002Fblob\u002Fmaster\u002Fdemos\u002Fcwmdbg.ipynb) | 24.09.2025 |\n| Qwen3-Omni | Single multimodal model that, for the first time, maintains state-of-the-art performance across text, image, audio, and video without any degradation relative to single-modal counterparts | \u003Cul>\u003Cli>[Jin Xu](https:\u002F\u002Fjxu-thu.github.io)\u003C\u002Fli> \u003Cli>[Zhifang Guo](https:\u002F\u002Fopenreview.net\u002Fprofile?id=~Zhifang_Guo3)\u003C\u002Fli> \u003Cli>[Hangrui Hu](https:\u002F\u002Fgithub.com\u002Fhangruihu)\u003C\u002Fli> \u003Cli>[Yunfei Chu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=41QhCyYAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xiong Wang](https:\u002F\u002Fgithub.com\u002Fwangxiongts)\u003C\u002Fli> \u003Cli>[Jinzheng He](https:\u002F\u002Fgithub.com\u002FJinzhengHe)\u003C\u002Fli> \u003Cli>[Yuxuan Wang](https:\u002F\u002Fyuxuanw.me)\u003C\u002Fli> \u003Cli>[Xian Shi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=UCiUsSEAAAAJ)\u003C\u002Fli> \u003Cli>[Ting He](https:\u002F\u002Fgithub.com\u002Ftinghe)\u003C\u002Fli> \u003Cli>[Xinfa Zhu](https:\u002F\u002Fgithub.com\u002FXinfaZhu)\u003C\u002Fli> \u003Cli>[Yuanjun Lv](https:\u002F\u002Fgithub.com\u002FYuanjunLv)\u003C\u002Fli> \u003Cli>[Yongqi Wang](https:\u002F\u002Fgithub.com\u002FYongqiWang03)\u003C\u002Fli> \u003Cli>[Dake Guo](https:\u002F\u002Fgithub.com\u002FDakeGuo)\u003C\u002Fli> \u003Cli>[He Wang](https:\u002F\u002Fgithub.com\u002Fhwwang21)\u003C\u002Fli> \u003Cli>[Linhan Ma](https:\u002F\u002Fgithub.com\u002FLinhanMa)\u003C\u002Fli> \u003Cli>[Pei Zhang](https:\u002F\u002Fgithub.com\u002FPeiZhang98)\u003C\u002Fli> \u003Cli>[Xinyu Zhang](https:\u002F\u002Fgithub.com\u002FXinyuZhang97)\u003C\u002Fli> \u003Cli>[Hongkun Hao](https:\u002F\u002Fgithub.com\u002FHongkunHao)\u003C\u002Fli> \u003Cli>[Zishan Guo](https:\u002F\u002Fgithub.com\u002FZishanGuo)\u003C\u002Fli> \u003Cli>[Baosong Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=DE68IOIAAAAJ)\u003C\u002Fli> \u003Cli>[Bin Zhang](https:\u002F\u002Fgithub.com\u002Fbinzhaozhao)\u003C\u002Fli> \u003Cli>[Ziyang Ma](https:\u002F\u002Fgithub.com\u002FZiyangMa)\u003C\u002Fli> \u003Cli>[Xipin Wei](https:\u002F\u002Fgithub.com\u002FXipinWei)\u003C\u002Fli> \u003Cli>[Shuai Bai](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=iv2dcv8AAAAJ)\u003C\u002Fli> \u003Cli>[Keqin Chen](https:\u002F\u002Fgithub.com\u002Fkeqinc)\u003C\u002Fli> \u003Cli>[Xuejing Liu](https:\u002F\u002Fgithub.com\u002FXuejingL)\u003C\u002Fli> \u003Cli>[Peng Wang](https:\u002F\u002Fgithub.com\u002Fpengwang114)\u003C\u002Fli> \u003Cli>[Mingkun Yang](https:\u002F\u002Fgithub.com\u002FMingkunYang)\u003C\u002Fli> \u003Cli>[Dayiheng Liu](https:\u002F\u002Fdayihengliu.com)\u003C\u002Fli> \u003Cli>[Xingzhang Ren](https:\u002F\u002Fgithub.com\u002FXingzhangRen)\u003C\u002Fli> \u003Cli>[Bo Zheng](https:\u002F\u002Fgithub.com\u002FBakerBunker)\u003C\u002Fli> \u003Cli>[Rui Men](https:\u002F\u002Fgithub.com\u002FRuiMen)\u003C\u002Fli> \u003Cli>[Fan Zhou](https:\u002F\u002Fgithub.com\u002FFanZhou920)\u003C\u002Fli> \u003Cli>[Bowen Yu](https:\u002F\u002Fbowenyubig.github.io)\u003C\u002Fli> \u003Cli>[Jianxin Yang](https:\u002F\u002Fgithub.com\u002FJianxinYang)\u003C\u002Fli> \u003Cli>[Le Yu](https:\u002F\u002Fgithub.com\u002FYuLeAI)\u003C\u002Fli> \u003Cli>[Jingren Zhou](https:\u002F\u002Fwww.alibabagroup.com\u002Fen-US\u002Fnews\u002Fcto-jingren-zhou)\u003C\u002Fli> \u003Cli>[Junyang Lin](https:\u002F\u002Fgithub.com\u002Flinjunyang)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FQwenLM\u002FQwen3-Omni?style=social)](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen3-Omni) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2509.17765)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fqwen.ai\u002Fblog?id=65f766fc2dcba7905c1cb69cc4cab90e94126bf4&from=research.latest-advancements-list)\u003C\u002Fli>\u003Cli>[chat](https:\u002F\u002Fchat.qwen.ai\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FCV4E9rpNSD)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fqwenllm\u002Fqwen3-omni)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fhelp.aliyun.com\u002Fzh\u002Fmodel-studio\u002Fuser-guide\u002Fqwen-omni)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDao-AILab\u002Fflash-attention)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FQwen\u002FQwen3-Omni-Demo), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FQwen\u002FQwen3-Omni-Captioner-Demo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science-in-your-pocket\u002Fqwen3-omni-one-llm-for-text-images-audio-and-videos-aad51ea1a4e3)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FAlibaba_Qwen\u002Fstatus\u002F1970181599133344172)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_zdOrPju4_g), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0N8mif_OUlM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FCWSYLPJz8j0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FtRqEAN61qpk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FQwenLM\u002FQwen3-Omni\u002Fblob\u002Fmaster\u002Fcookbooks\u002Fimage_math.ipynb) | 22.09.2025 |\n| WMAR | Custom tokenizer-detokenizer finetuning procedure that improves RCC, and a complementary watermark synchronization layer | \u003Cul>\u003Cli>[Nikola Jovanović](https:\u002F\u002Fwww.sri.inf.ethz.ch\u002Fpeople\u002Fnikola)\u003C\u002Fli> \u003Cli>[Ismail Labiad](https:\u002F\u002Fgithub.com\u002Filabiad)\u003C\u002Fli> \u003Cli>[Tomáš Souček](https:\u002F\u002Fgithub.com\u002FsoCzech)\u003C\u002Fli> \u003Cli>[Martin Vechev](https:\u002F\u002Fwww.sri.inf.ethz.ch\u002Fpeople\u002Fmartin)\u003C\u002Fli> \u003Cli>[Pierre Fernandez](https:\u002F\u002Fpierrefdz.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fwmar?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fwmar) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.16349)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002FDiffPure), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fchameleon), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbytedance\u002F1d-tokenizer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmehrdadsaberi\u002Fwatermark_robustness)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fwmar\u002Fblob\u002Fmain\u002Fnotebooks\u002Fcolab.ipynb) | 19.09.2025 |\n| LIMIT | On the Theoretical Limitations of Embedding-Based Retrieval | \u003Cul>\u003Cli>[Orion Weller](https:\u002F\u002Forionweller.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michael Boratko](https:\u002F\u002Fwww.mboratko.com\u002F)\u003C\u002Fli> \u003Cli>[Iftekhar Naim](https:\u002F\u002Fgithub.com\u002Fiftekharnaim)\u003C\u002Fli> \u003Cli>[Jinhyuk Lee](https:\u002F\u002Fjhyuklee.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Flimit?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Flimit) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.21038)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fembeddings-benchmark\u002Fmteb)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Forionweller\u002FLIMIT), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Forionweller\u002FLIMIT-small)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fnoailabs.medium.com\u002Fon-the-theoretical-limitations-of-embedding-based-retrieval-new-research-paper-c70dc3edc817)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-deepmind\u002Flimit\u002Fblob\u002Fmaster\u002Fcode\u002Fgenerate_limit_dataset.ipynb) | 27.08.2025 |\n| DINOv3 | Produces high-quality dense features that achieve outstanding performance on various vision tasks, significantly surpassing previous self- and weakly-supervised foundation models | \u003Cul>\u003Cli>[Oriane Siméoni](https:\u002F\u002Forlanesimeoni.github.io\u002F)\u003C\u002Fli> \u003Cli>[Huy Vo](https:\u002F\u002Fhuyvvo.github.io\u002F)\u003C\u002Fli> \u003Cli>[Maximilian Seitzer](https:\u002F\u002Fgithub.com\u002FSeitzerM)\u003C\u002Fli> \u003Cli>[Federico Baldassarre](https:\u002F\u002Ffedebal.dev\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Maxime Oquab](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5vteYV8AAAAJ)\u003C\u002Fli> \u003Cli>[Cijo Jose](https:\u002F\u002Fgithub.com\u002Fcijojose)\u003C\u002Fli> \u003Cli>[Vasil Khalidov](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=tjazz3AAAAAJ)\u003C\u002Fli> \u003Cli>[Marc Szafraniec](https:\u002F\u002Fgithub.com\u002FMarcSzafraniec)\u003C\u002Fli> \u003Cli>[Seungeun Yi](https:\u002F\u002Fyi-seungeun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michaël Ramamonjisoa](https:\u002F\u002Fmichaelramamonjisoa.github.io\u002F)\u003C\u002Fli> \u003Cli>[Francisco Massa](https:\u002F\u002Fgithub.com\u002Ffmassa)\u003C\u002Fli> \u003Cli>[Daniel Haziza](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2eSKdFMAAAAJ)\u003C\u002Fli> \u003Cli>[Luca Wehrstedt](https:\u002F\u002Flucawehrstedt.info\u002F)\u003C\u002Fli> \u003Cli>[Jianyuan Wang](https:\u002F\u002Fjianyuanwang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Timothée Darcet](https:\u002F\u002Fgithub.com\u002FTimDarcet)\u003C\u002Fli> \u003Cli>[Théo Moutakanni](https:\u002F\u002Fgithub.com\u002FTheoMoutak)\u003C\u002Fli> \u003Cli>[Leonel Sentana](https:\u002F\u002Fleosentana.github.io\u002F)\u003C\u002Fli> \u003Cli>[Claire Roberts](https:\u002F\u002Fclaire-roberts.github.io\u002F)\u003C\u002Fli> \u003Cli>[Andrea Vedaldi](https:\u002F\u002Fandrea-vedaldi.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jamie Tolan](https:\u002F\u002Fjamietolan.github.io\u002F)\u003C\u002Fli> \u003Cli>[John Brandt](https:\u002F\u002Fjohnbrandt.github.io\u002F)\u003C\u002Fli> \u003Cli>[Camille Couprie](https:\u002F\u002Fcamillecouprie.github.io\u002F)\u003C\u002Fli> \u003Cli>[Julien Mairal](http:\u002F\u002Fthoth.inrialpes.fr\u002Fpeople\u002Fmairal\u002F)\u003C\u002Fli> \u003Cli>[Hervé Jégou](https:\u002F\u002Fgithub.com\u002Fjegou)\u003C\u002Fli> \u003Cli>[Patrick Labatut](https:\u002F\u002Fgithub.com\u002Fpatricklabatut)\u003C\u002Fli> \u003Cli>[Piotr Bojanowski](https:\u002F\u002Fgithub.com\u002Fpiotr-bojanowski)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fdinov3?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdinov3) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2508.10104), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.16334)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Ffacebook\u002Fdinov3-68924841bd6b561778e31009), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Fdinov3)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fdatasets\u002Fnikhil7280\u002Fcoco-image-caption)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science-in-your-pocket\u002Fmeta-dino-v3-the-ultimate-vision-ai-for-every-image-task-cf5ffc30a221)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fdinov3\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fdinov3-self-supervised-vision-model\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresources\u002Fmodels-and-libraries\u002Fdinov3-downloads\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-eOYWK6m3i8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fdinov3\u002Fblob\u002Fmain\u002Fnotebooks\u002Fsegmentation_tracking.ipynb) | 14.08.2025 |\n| Hogwild! Inference | Run LLM \"workers\" in parallel, allowing them to synchronize via a concurrently-updated attention cache and prompt these workers to decide how best to collaborate | \u003Cul>\u003Cli>[Gleb Rodionov](https:\u002F\u002Fgithub.com\u002Feqimp)\u003C\u002Fli> \u003Cli>[Roman Garipov](https:\u002F\u002Fgithub.com\u002Fgaripovroma)\u003C\u002Fli> \u003Cli>[Alina Shutova](https:\u002F\u002Fgithub.com\u002Fgoodevening13)\u003C\u002Fli> \u003Cli>[George Yakushev](https:\u002F\u002Fgithub.com\u002FMr-DarkTesla)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Vage Egiazarian](https:\u002F\u002Fgithub.com\u002FVahe1994)\u003C\u002Fli> \u003Cli>[Anton Sinitsin](https:\u002F\u002Fgithub.com\u002Fxtinkt)\u003C\u002Fli> \u003Cli>[Denis Kuznedelev](https:\u002F\u002Fgithub.com\u002FGodofnothing)\u003C\u002Fli> \u003Cli>[Dan Alistarh](https:\u002F\u002Fgithub.com\u002Fdalistarh)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Feqimp\u002Fhogwild_llm?style=social)](https:\u002F\u002Fgithub.com\u002Feqimp\u002Fhogwild_llm) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2504.06261), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.11903), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2406.04692), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.15337)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FQwen\u002FQwQ-32B)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fllama-4-multimodal-intelligence\u002F)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Feqimp.github.io\u002Fhogwild_llm\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLocalLLaMA\u002Fcomments\u002F1jv7x6l\u002Fhogwild_inference_parallel_llm_generation_via\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Feqimp\u002Fhogwild_llm\u002Fblob\u002Fmain\u002Fcolab_example.ipynb) | 15.07.2025 |\n| Grounding DINO | Marrying DINO with Grounded Pre-Training for Open-Set Object Detection | \u003Cul>\u003Cli>[Shilong Liu](https:\u002F\u002Fgithub.com\u002FSlongLiu)\u003C\u002Fli> \u003Cli>[Zhaoyang Zeng](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=U_cvvUwAAAAJ)\u003C\u002Fli> \u003Cli>[Tianhe Ren](https:\u002F\u002Frentainhe.github.io\u002F)\u003C\u002Fli> \u003Cli>[Feng Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ybRe9GcAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Hao Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=B8hPxMQAAAAJ)\u003C\u002Fli> \u003Cli>[Jie Yang](https:\u002F\u002Fyangjie-cv.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chunyuan Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Zd7WmXUAAAAJ)\u003C\u002Fli> \u003Cli>[Jianwei Yang](https:\u002F\u002Fjwyang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hang Su](https:\u002F\u002Fwww.suhangss.me\u002F)\u003C\u002Fli> \u003Cli>[Jun Zhu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=axsP38wAAAAJ)\u003C\u002Fli> \u003Cli>[Lei Zhang](https:\u002F\u002Fwww.leizhang.org\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FIDEA-Research\u002FGroundingDINO?style=social)](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002FGroundingDINO) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.05499)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002FDINO), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FUX-Decoder\u002FSemantic-SAM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOptimalScale\u002FDetGPT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002FOpenSeeD), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FUX-Decoder\u002FSegment-Everything-Everywhere-All-At-Once), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FX-Decoder\u002Ftree\u002Fxgpt), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002Fdetrex)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fzero-shot-object-detection-on-mscoco?p=grounding-dino-marrying-dino-with-grounded), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fzero-shot-object-detection-on-odinw?p=grounding-dino-marrying-dino-with-grounded), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fobject-detection-on-coco-minival?p=grounding-dino-marrying-dino-with-grounded), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fobject-detection-on-coco?p=grounding-dino-marrying-dino-with-grounded)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FwxWDt5UiwY8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FcMa77r3YrDk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FC4NqaRBz_Kw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FoEQYStnF2l8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Froboflow-ai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnotebooks\u002Fzero-shot-object-detection-with-grounding-dino.ipynb) | 10.07.2025 |\n| Hunyuan | Open-source large language model built on a fine-grained Mixture-of-Experts architecture | [manayang](https:\u002F\u002Fgithub.com\u002FManaEstras) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTencent-Hunyuan\u002FHunyuan-A13B?style=social)](https:\u002F\u002Fgithub.com\u002FTencent-Hunyuan\u002FHunyuan-A13B) \u003Cul>\u003Cli>[api](https:\u002F\u002Fcloud.tencent.com\u002Fproduct\u002Ftclm)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>]()\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FbsPcMEtV7v)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fhunyuaninfer\u002Fhunyuan-a13b\u002Ftags), [\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fhunyuaninfer\u002Fhunyuan-large\u002Ftags)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTencent\u002FAngelSlim)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ftencent\u002FHunyuan-A13B-Instruct)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLocalLLaMA\u002Fcomments\u002F1o22v1b\u002Fwhat_are_your_thoughts_on\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FTencentHunyuan\u002Fstatus\u002F1938525874904801490)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fhunyuan.tencent.com\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FH7JHky8Yh9Y), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYiQv6Av0jQg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQmRmcn_zOnU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBGtkuVKpVUA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdatawhalechina\u002Fself-llm\u002Fblob\u002Fmaster\u002Fmodels\u002FHunyuan-A13B-Instruct\u002F05-Hunyuan-A13B-Instruct-LoRA.ipynb) | 01.07.2025 |\n| Whisper | Automatic speech recognition system trained on 680,000 hours of multilingual and multitask supervised data collected from the web | \u003Cul>\u003Cli>[Alec Radford](http:\u002F\u002Fnewmu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jong Wook Kim](https:\u002F\u002Fjongwook.kim\u002F)\u003C\u002Fli> \u003Cli>[Tao Xu](https:\u002F\u002Fgithub.com\u002Fbayesian)\u003C\u002Fli> \u003Cli>[Greg Brockman](https:\u002F\u002Fgregbrockman.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Christine McLeavey](http:\u002F\u002Fchristinemcleavey.com\u002F)\u003C\u002Fli> \u003Cli>[Ilya Sutskever](http:\u002F\u002Fwww.cs.toronto.edu\u002F~ilya\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fopenai\u002Fwhisper?style=social)](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fwhisper) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.04356)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fopenai.com\u002Fresearch\u002Fwhisper)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkkroening\u002Fffmpeg-python)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FOCBZtgQGt1I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8SQV-B83tPU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FnE5iVtwKerA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fopenai\u002Fwhisper\u002Fblob\u002Fmaster\u002Fnotebooks\u002FLibriSpeech.ipynb) | 26.06.2025 |\n| IT³ | Idempotent Test-Time Training, approach that enables on-the-fly adaptation to distribution shifts using only the current test instance, without any auxiliary task design | \u003Cul>\u003Cli>[Nikita Durasov](https:\u002F\u002Fwww.norange.io\u002Fabout\u002F)\u003C\u002Fli> \u003Cli>[Assaf Shocher](https:\u002F\u002Fassafshocher.github.io\u002F)\u003C\u002Fli> \u003Cli>[Doruk Oner](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ESA2CsAAAAAJ)\u003C\u002Fli> \u003Cli>[Gal Chechik](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Fgal-chechik)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Alexei Efros](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~efros\u002F)\u003C\u002Fli> \u003Cli>[Pascal Fua](https:\u002F\u002Fpeople.epfl.ch\u002Fpascal.fua\u002Fbio)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fnikitadurasov\u002Fittt?style=social)](https:\u002F\u002Fgithub.com\u002Fnikitadurasov\u002Fittt) \u003Cul>\u003Cli>[ICML](https:\u002F\u002Ficml.cc\u002Fvirtual\u002F2025\u002Fposter\u002F45551)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.04201)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnikitadurasov\u002Ftorch-ttt)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.norange.io\u002Fprojects\u002Fittt\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Ftorch-ttt\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FeKGKpN8fFRM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fn5opIsl9WRA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fnikitadurasov\u002Fittt\u002Fblob\u002Fmain\u002Fexps\u002Fmnist\u002Fit3_torch_ttt.ipynb) | 25.06.2025 |\n| AlphaEvolve | Evolutionary coding agent that substantially enhances capabilities of state-of-the-art LLMs on highly challenging tasks such as tackling open scientific problems or optimizing critical pieces of computational infrastructure | \u003Cul>\u003Cli>[Alexander Novikov](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jMUkLqwAAAAJ)\u003C\u002Fli> \u003Cli>[Ngân Vu](https:\u002F\u002Flinkedin.com\u002Fin\u002Fthuynganvu)\u003C\u002Fli> \u003Cli>[Marvin Eisenberger](https:\u002F\u002Fcvg.cit.tum.de\u002Fmembers\u002Feisenber)\u003C\u002Fli> \u003Cli>[Emilien Dupont](https:\u002F\u002Femiliendupont.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Po-Sen Huang](https:\u002F\u002Fposenhuang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Adam Wagner](https:\u002F\u002Fzawagner22.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sergey Shirobokov](https:\u002F\u002Fgithub.com\u002Fshir994)\u003C\u002Fli> \u003Cli>[Borislav Kozlovskii](https:\u002F\u002Flinkedin.com\u002Fin\u002Fborislav-kozlovskii-63801b192)\u003C\u002Fli> \u003Cli>[Francisco Ruiz](https:\u002F\u002Ffranrruiz.github.io\u002F)\u003C\u002Fli> \u003Cli>[Abbas Mehrabian](https:\u002F\u002Fabbasmehrabian.com\u002F)\u003C\u002Fli> \u003Cli>[Pawan Kumar](https:\u002F\u002Fwww.ellis.ox.ac.uk\u002Fpeople\u002Fm-pawan-kumar)\u003C\u002Fli> \u003Cli>[Abigail See](https:\u002F\u002Fcs.stanford.edu\u002Fpeople\u002Fabisee\u002F)\u003C\u002Fli> \u003Cli>[Swarat Chaudhuri](https:\u002F\u002Fwww.cs.utexas.edu\u002F~swarat\u002F)\u003C\u002Fli> \u003Cli>[George Holland](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fg-aracil-holland)\u003C\u002Fli> \u003Cli>[Alex Davies](https:\u002F\u002Fwww.alexdavies.net\u002F)\u003C\u002Fli> \u003Cli>[Sebastian Nowozin](https:\u002F\u002Fwww.nowozin.net\u002Fsebastian\u002F)\u003C\u002Fli> \u003Cli>[Pushmeet Kohli](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FPushmeet_Kohli)\u003C\u002Fli> \u003Cli>[Matej Balog](http:\u002F\u002Fmatejbalog.eu\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Falphaevolve_results?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Falphaevolve_results) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fstorage.googleapis.com\u002Fdeepmind-media\u002FDeepMind.com\u002FBlog\u002Falphaevolve-a-gemini-powered-coding-agent-for-designing-advanced-algorithms\u002FAlphaEvolve.pdf), [\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fdiscover\u002Fblog\u002Falphaevolve-a-gemini-powered-coding-agent-for-designing-advanced-algorithms\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-alphaevolve-a-coding-agent-for-scientific-and-algorithmic-discovery-5732a876c2e2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fsingularity\u002Fcomments\u002F1kmhti8\u002Fdeepmind_introduces_alphaevolve_a_geminipowered\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FAlphaEvolve)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjCTvblRXnzg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FsGCmu7YKgPA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FT0eWBlFhFzc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDYyK76ZOUJU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FEMoiremdiA8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0-MA3jgYMMg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fx1FFLzTX-Kg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRH4hAgvYSzg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-deepmind\u002Falphaevolve_results\u002Fblob\u002Fmain\u002Fmathematical_results.ipynb) | 17.06.2025 |\n| V-JEPA 2 | Self-supervised approach that combines internet-scale video data with a small amount of interaction data, to develop models capable of understanding, predicting, and planning in the physical world | [FAIR](https:\u002F\u002Fai.meta.com\u002Fresearch\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fvjepa2?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fvjepa2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2506.09985)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fandlukyane.com\u002Fblog\u002Fpaper-review-vjepa2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Ffacebook\u002Fv-jepa-2-6841bad8413014e185b497a6)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-v-jepa-2-self-supervised-video-models-enable-understanding-prediction-and-planning-28410d8a1c6b)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fv-jepa-2-world-model-benchmarks)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FijEmo75zqZc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FsMO3BAPBLG4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fvjepa2\u002Fblob\u002Fmain\u002Fnotebooks\u002Fvjepa2_demo.ipynb) | 11.06.2025 |\n| TimesFM | Time-series foundation model for forecasting whose out-of-the-box zero-shot performance on a variety of public datasets comes close to the accuracy of state-of-the-art supervised forecasting models for each individual dataset | \u003Cul>\u003Cli>[Abhimanyu Das](https:\u002F\u002Fgithub.com\u002Fabhimanyudas747)\u003C\u002Fli> \u003Cli>[Weihao Kong](https:\u002F\u002Fweihaokong.github.io\u002F)\u003C\u002Fli> \u003Cli>[Rajat Sen](https:\u002F\u002Fgithub.com\u002Frajatsen91)\u003C\u002Fli> \u003Cli>[Yichen Zhou](https:\u002F\u002Fgithub.com\u002Fsiriuz42)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Ftimesfm?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Ftimesfm) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.10688)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fresearch.google\u002Fblog\u002Fa-decoder-only-foundation-model-for-time-series-forecasting\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Ftimesfm), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FSalesforce\u002FGIFT-Eval), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Fgoogle\u002Ftimesfm-release-66e4be5fdb56e960c1e482a6)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Ftimesfm\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_2SWu1SOcG0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYkzRP3xnMwc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F265Mpaj8O1U), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FOhEAS5oBcco)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FNielsRogge\u002FTransformers-Tutorials\u002Fblob\u002Fmaster\u002FTimesFM\u002FFine_tune_TimesFM_on_a_custom_dataset.ipynb) | 26.05.2025 |\n| Qwen2.5-Omni | End-to-end multimodal model designed to perceive diverse modalities, including text, images, audio, and video, while simultaneously generating text and natural speech responses in a streaming manner | \u003Cul>\u003Cli>[Jin Xu](https:\u002F\u002Fjxu-thu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhifang Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=LHzzlAgAAAAJ)\u003C\u002Fli> \u003Cli>[Jinzheng He](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=tsQVykcAAAAJ)\u003C\u002Fli> \u003Cli>[Ting He](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Yxat_NMAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Shuai Bai](https:\u002F\u002Fgithub.com\u002FShuaiBai623)\u003C\u002Fli> \u003Cli>[Keqin Chen](https:\u002F\u002Fgithub.com\u002FOliverChen0602)\u003C\u002Fli> \u003Cli>[Jialin Wang](https:\u002F\u002Fgithub.com\u002FJialinWangPKU)\u003C\u002Fli> \u003Cli>[Yang Fan](https:\u002F\u002Fgithub.com\u002Ffyabc)\u003C\u002Fli> \u003Cli>[Peng Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=7fjqA0YAAAAJ)\u003C\u002Fli> \u003Cli>[Bin Zhang](https:\u002F\u002Fzhangchbin.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xiong Wang](https:\u002F\u002Fgithub.com\u002Fwangxiongts)\u003C\u002Fli> \u003Cli>[Yunfei Chu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=41QhCyYAAAAJ)\u003C\u002Fli> \u003Cli>[Junyang Lin](https:\u002F\u002Fjustinlin610.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FQwenLM\u002FQwen2.5-Omni?style=social)](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen2.5-Omni) \u003Cul>\u003Cli>[API](https:\u002F\u002Fhelp.aliyun.com\u002Fzh\u002Fmodel-studio\u002Fuser-guide\u002Fqwen-omni)\u003C\u002Fli>\u003Cli>[Chat](https:\u002F\u002Fchat.qwen.ai\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2503.20215)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fqwenlm.github.io\u002Fblog\u002Fqwen2.5-omni\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FCV4E9rpNSD)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fqwenllm\u002Fqwen-omni)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDao-AILab\u002Fflash-attention)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002FQwen\u002Fqwen25-omni-67de1e5f0f9464dc6314b36e), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FQwen\u002FQwen2.5-Omni-7B-Demo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science-in-your-pocket\u002Fqwen2-5-omni-7b-input-anything-output-text-and-audio-llm-87c050716172)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyKcANdkRuNI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-JY1wcRRXMA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVez5TyC5YTE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5BqFvIm0joU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FQwenLM\u002FQwen2.5-Omni\u002Fblob\u002Fmain\u002Fcookbooks\u002Fomni_chatting_for_math.ipynb) | 29.04.2025 |\n| EAT | Emotional Adaptation for Audio-driven Talking-head method, which transforms emotion-agnostic talking-head models into emotion-controllable ones in a cost-effective and efficient manner through parameter-efficient adaptations | \u003Cul>\u003Cli>[Yuan Gan](https:\u002F\u002Fyuangan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zongxin Yang](https:\u002F\u002Fz-x-yang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xihang Yue](https:\u002F\u002Fgithub.com\u002Fyuexihang)\u003C\u002Fli> \u003Cli>[Lingyun Sun](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=zzW8d-wAAAAJ)\u003C\u002Fli> \u003Cli>[Yi Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RMSuNFwAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2fe6fd6c5a7e.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.02069) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyuangan\u002FEAT_code?style=social)](https:\u002F\u002Fgithub.com\u002Fyuangan\u002FEAT_code) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.04946)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyuangan\u002Fevaluation_eat), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYudongGuo\u002FAD-NeRF), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjixinya\u002FEAMM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhanglonghao1992\u002FOne-Shot_Free-View_Neural_Talking_Head_Synthesis), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFuxiVirtualHuman\u002FAAAI22-one-shot-talking-face), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHangz-nju-cuhk\u002FTalking-Face_PC-AVS), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fvid2vid)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyuangan.github.io\u002Feat\u002F)\u003C\u002Fli>\u003Cli>[video](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1sAoplzY4b6JCW0JQHf_HKEL5luuWuGAk\u002Fview)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Flp2nSLZp-88)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F133hwDHzsfRYl-nQCUQxJGjcXa5Fae22Z) | 22.04.2025 |\n| Moshi | Speech-text foundation model and full-duplex spoken dialogue framework | \u003Cul>\u003Cli>[Alexandre Défossez](https:\u002F\u002Fgithub.com\u002Fadefossez)\u003C\u002Fli> \u003Cli>[Laurent Mazaré](http:\u002F\u002Flaurentmazare.github.io)\u003C\u002Fli> \u003Cli>[Manu Orsini](https:\u002F\u002Fwww.facebook.com\u002Forsini9\u002F)\u003C\u002Fli> \u003Cli>[Amélie Royer](https:\u002F\u002Ffr.linkedin.com\u002Fin\u002Famélie-royer-aa26081a)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Patrick Pérez](https:\u002F\u002Fgithub.com\u002Fpppjer)\u003C\u002Fli> \u003Cli>[Hervé Jégou](https:\u002F\u002Fgithub.com\u002Fjerfju)\u003C\u002Fli> \u003Cli>[Edouard Grave](https:\u002F\u002Ffr.linkedin.com\u002Fin\u002Fedouard-grave-63099823)\u003C\u002Fli> \u003Cli>[Neil Zeghidour](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=p_aUHWgAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fkyutai-labs\u002Fmoshi?style=social)](https:\u002F\u002Fgithub.com\u002Fkyutai-labs\u002Fmoshi) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.00037), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2410.00037), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.03312), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2110.13900), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.14090)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fmoshi.chat\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkyutai-labs\u002Fmoshi-finetune), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZhangXInFD\u002FSpeechTokenizer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhaoheliu\u002FSemantiCodec-inference), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fencodec)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Fkyutai\u002Fmoshi-v01-release-66eaeaf3302bef6bd9ad7acd)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fmoshi\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1fmmm18\u002Fmoshi_a_speechtext_foundation_model_for_real_time\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0_c3bw_x6uU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FK1TbRsgwWd4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPO4EO7kUUQQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FedToLqBw_K8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDv31P1aTVOs)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fkyutai-labs\u002Fmoshi\u002Fblob\u002Fmaster\u002Fmoshi\u002Fdemo_moshi.ipynb) | 31.03.2025 |\n| BiRefNet | Bilateral reference framework for high-resolution dichotomous image segmentation | \u003Cul>\u003Cli>[Peng Zheng](https:\u002F\u002Fzhengpeng7.github.io\u002Fabout\u002F)\u003C\u002Fli> \u003Cli>[Dehong Gao](https:\u002F\u002Fteacher.nwpu.edu.cn\u002Fdehonggao)\u003C\u002Fli> \u003Cli>[Deng-Ping Fan](https:\u002F\u002Fdengpingfan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Li Liu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=9cMQrVsAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jorma Laaksonen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=qQP6WXIAAAAJ)\u003C\u002Fli> \u003Cli>[Wanli Ouyang](https:\u002F\u002Fwlouyang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nicu Sebe](https:\u002F\u002Fdisi.unitn.it\u002F~sebe\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_b8af1a40a030.png)](https:\u002F\u002Fdoi.org\u002F10.26599\u002FAIR.2024.9150038) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FZhengPeng7\u002FBiRefNet?style=social)](https:\u002F\u002Fgithub.com\u002FZhengPeng7\u002FBiRefNet) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.03407), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2302.14485)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002Fd9NN5sgFrq)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FKazuhito00\u002FBiRefNet-ONNX-Sample), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZHO-ZHO-ZHO\u002FComfyUI-BiRefNet-ZHO), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fviperyl\u002FComfyUI-BiRefNet)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FZhengPeng7\u002FBiRefNet_demo), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FZhengPeng7\u002FBiRefNet)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.birefnet.top\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fdichotomous-image-segmentation-on-dis-te1?p=bilateral-reference-for-high-resolution), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fcamouflaged-object-segmentation-on-cod?p=bilateral-reference-for-high-resolution), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Frgb-salient-object-detection-on-davis-s?p=bilateral-reference-for-high-resolution)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1B6aKZ3ekcvKMkSBn0N5mCASLUYMp0whK) | 24.03.2025 |\n| ESM | Evolutionary Scale Modeling: Pretrained language models for proteins | \u003Cul>\u003Cli>[Zeming Lin](https:\u002F\u002Fresearch.facebook.com\u002Fpeople\u002Flin-zeming\u002F)\u003C\u002Fli> \u003Cli>[Roshan Rao](https:\u002F\u002Frmrao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Brian Hie](https:\u002F\u002Fbrianhie.com\u002F)\u003C\u002Fli> \u003Cli>[Zhongkai Zhu](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fzhongkai-zhu-03a27424)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Allan dos Santos Costa](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Zb4RsFsAAAAJ)\u003C\u002Fli> \u003Cli>[Maryam Fazel-Zarandi](https:\u002F\u002Fwww.maryamfazel.com\u002F)\u003C\u002Fli> \u003Cli>[Tom Sercu](https:\u002F\u002Ftom.sercu.me\u002F)\u003C\u002Fli> \u003Cli>[Salvatore Candido](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=BDgbhmEAAAAJ)\u003C\u002Fli> \u003Cli>[Alexander Rives](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vqb78-gAAAAJ)\u003C\u002Fli> \u003Cli>[Joshua Meier](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2M0OltAAAAAJ)\u003C\u002Fli> \u003Cli>[Robert Verkuil](https:\u002F\u002Fdblp.org\u002Fpid\u002F296\u002F8930.html)\u003C\u002Fli> \u003Cli>[Jason Liu](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fliujiayi\u002F)\u003C\u002Fli> \u003Cli>[Chloe Hsu](https:\u002F\u002Fchloe-hsu.com\u002F)\u003C\u002Fli> \u003Cli>[Adam Lerer](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Ad6O4-0AAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9220bdc4ae8c.png)](https:\u002F\u002Fdoi.org\u002F10.1101\u002F622803) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fesm?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fesm) \u003Cul>\u003Cli>[ESM Atlas](https:\u002F\u002Fesmatlas.com\u002F)\u003C\u002Fli>\u003Cli>[FSDP](https:\u002F\u002Ffairscale.readthedocs.io\u002Fen\u002Fstable\u002Fapi\u002Fnn\u002Ffsdp.html)\u003C\u002Fli>\u003Cli>[ICML](https:\u002F\u002Fproceedings.mlr.press\u002Fv139\u002Frao21a.html)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fftp.uniprot.org\u002Fpub\u002Fdatabases\u002Funiprot\u002Fprevious_releases\u002Frelease-2018_03\u002Funiref\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsokrypton\u002FColabFold)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Fesm)\u003C\u002Fli>\u003Cli>[paper](https:\u002F\u002Fdoi.org\u002F10.1101\u002F2022.07.20.500902), [paper](https:\u002F\u002Fdoi.org\u002F10.1101\u002F2021.07.09.450648), [paper](https:\u002F\u002Fdoi.org\u002F10.1101\u002F2022.04.10.487779), [paper](https:\u002F\u002Fdoi.org\u002F10.1101\u002F2022.12.21.521521)\u003C\u002Fli>\u003Cli>[pubmed](https:\u002F\u002Fpubmed.ncbi.nlm.nih.gov\u002F33876751\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FN-eisTvUYrk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGHoE4VkDehY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsokrypton\u002FColabFold\u002Fblob\u002Fmain\u002FESMFold.ipynb) | 21.03.2025 |\n| Video Seal | Comprehensive framework for neural video watermarking and a competitive open-sourced model | \u003Cul>\u003Cli>[Pierre Fernandez](https:\u002F\u002Fpierrefdz.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hady Elsahar](https:\u002F\u002Fwww.hadyelsahar.io\u002F)\u003C\u002Fli> \u003Cli>[Zeki Yalniz](https:\u002F\u002Fai.meta.com\u002Fpeople\u002F274060152308398\u002Fi-zeki-yalniz\u002F)\u003C\u002Fli> \u003Cli>[Alexandre Mourachko](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=OD9-erYAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fvideoseal?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fvideoseal) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.09492), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2510.20468)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Faidemos.meta.com\u002Fvideoseal)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresearch\u002Fpublications\u002Fvideo-seal-open-and-efficient-video-watermarking\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fvideoseal\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fvideoseal\u002Fblob\u002Fmain\u002Fnotebooks\u002Fcolab.ipynb) | 17.03.2025 |\n| SigLIP 2 | Family of new multilingual vision-language encoders that build on the success of the original SigLIP | \u003Cul>\u003Cli>[Michael Tschannen](https:\u002F\u002Fmitscha.github.io\u002F)\u003C\u002Fli> \u003Cli>[Alexey Gritsenko](https:\u002F\u002Fgithub.com\u002FAlexeyG)\u003C\u002Fli> \u003Cli>[Xiao Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ukyXqzMAAAAJ)\u003C\u002Fli> \u003Cli>[Muhammad Ferjad Naeem](https:\u002F\u002Fferjad.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ibrahim Alabdulmohsin](https:\u002F\u002Fibomohsin.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nikhil Parthasarathy](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=X9mO4ckAAAAJ)\u003C\u002Fli> \u003Cli>[Talfan Evans](https:\u002F\u002Ftalfanevans.co.uk\u002F)\u003C\u002Fli> \u003Cli>[Lucas Beyer](http:\u002F\u002Flucasb.eyer.be)\u003C\u002Fli> \u003Cli>[Ye Xia](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QQhJ1pAAAAAJ)\u003C\u002Fli> \u003Cli>[Basil Mustafa](https:\u002F\u002Fwww.basilmustafa.com\u002F)\u003C\u002Fli> \u003Cli>[Olivier Hénaff](https:\u002F\u002Fwww.olivierhenaff.com\u002F)\u003C\u002Fli> \u003Cli>[Jeremiah Harmsen](https:\u002F\u002Fresearch.google\u002Fpeople\u002Fjeremiahharmsen)\u003C\u002Fli> \u003Cli>[Andreas Steiner](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vIZeAu4AAAAJ)\u003C\u002Fli> \u003Cli>[Xiaohua Zhai](https:\u002F\u002Fsites.google.com\u002Fview\u002Fxzhai)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fbig_vision?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fbig_vision\u002Fblob\u002Fmain\u002Fbig_vision\u002Fmodels\u002Fvit.py) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2502.14786), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.15343)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fbig_vision\u002Fblob\u002Fmain\u002Fbig_vision\u002Fmodels\u002Fproj\u002Fimage_text\u002Ftwo_towers.py), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fbig_vision\u002Fblob\u002Fmain\u002Fbig_vision\u002Fmodels\u002Fproj\u002Fimage_text\u002Fnaflex_vit.py), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fbig_vision\u002Fblob\u002Fmain\u002Fbig_vision\u002Fpp\u002Fproj\u002Fimage_text\u002Fops_naflex.py)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Fgoogle\u002Fsiglip2-67b5dcef38c175486e240107)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-siglip-2-multilingual-vision-language-encoders-with-improved-semantic-understanding-b7b578002adc), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fritvik19.medium.com\u002Fpapers-explained-320-siglip-2-dba08ff09559)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fr8y7gXIpb4A)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Fbig_vision\u002Fblob\u002Fmain\u002Fbig_vision\u002Fconfigs\u002Fproj\u002Fimage_text\u002FSigLIP2_demo.ipynb) | 17.03.2025 |\n| DeepLabCut | Efficient method for markerless pose estimation based on transfer learning with deep neural networks that achieves excellent results with minimal training data | \u003Cul>\u003Cli>[Alexander Mathis](https:\u002F\u002Fgithub.com\u002FAlexEMG)\u003C\u002Fli> \u003Cli>[Pranav Mamidanna](https:\u002F\u002Fpranavm19.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kevin Cury](https:\u002F\u002Fkevincury.com\u002F)\u003C\u002Fli> \u003Cli>[Taiga Abe](https:\u002F\u002Fcellistigs.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Venkatesh Murthy](https:\u002F\u002Fgithub.com\u002Fvenkateshnmurthy)\u003C\u002Fli> \u003Cli>[Mackenzie Mathis](https:\u002F\u002Fgithub.com\u002FMMathisLab)\u003C\u002Fli> \u003Cli>[Matthias Bethge](https:\u002F\u002Fbethgelab.org\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_e9c6ba8d4911.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41593-018-0209-y) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FDeepLabCut\u002FDeepLabCut?style=social)](https:\u002F\u002Fgithub.com\u002FDeepLabCut\u002FDeepLabCut) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1605.03170), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.03142), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.11229), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2009.00564), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.13868), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.13868)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fdeeplabcut\u002Fdeeplabcut)\u003C\u002Fli>\u003Cli>[forum](https:\u002F\u002Fforum.image.sc\u002Ftag\u002Fdeeplabcut)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDeepLabCut\u002FDLCutils), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDeepLabCut\u002FDeepLabCut-Workshop-Materials)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@cziscience\u002Fhow-open-source-software-contributors-are-accelerating-biomedicine-1a5f50f6846a)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fdeeplabcut\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002FDeepLabCut)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fwww.deeplabcut.org\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002F@deeplabcut7702), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuWZu3rnj-kQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTeb5r2TNAYs)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FDeepLabCut\u002FDeepLabCut\u002Fblob\u002Fmaster\u002Fexamples\u002FCOLAB\u002FCOLAB_maDLC_TrainNetwork_VideoAnalysis.ipynb) | 28.02.2025 |\n| STAR | Spatial Temporal Augmentation with T2V models for Real-world video super-resolution, a novel approach that leverages T2V models for real-world video super-resolution, achieving realistic spatial details and robust temporal consistency | \u003Cul>\u003Cli>[Rui Xie](https:\u002F\u002Fgithub.com\u002FCSRuiXie)\u003C\u002Fli> \u003Cli>[Yinhong Liu](https:\u002F\u002Fgithub.com\u002Fyhliu04)\u003C\u002Fli> \u003Cli>[Penghao Zhou](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=yWq1Fd4AAAAJ)\u003C\u002Fli> \u003Cli>[Chen Zhao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Uhp3JKgAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jun Zhou](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=w03CHFwAAAAJ)\u003C\u002Fli> \u003Cli>[Kai Zhang](https:\u002F\u002Fcszn.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhenyu Zhang](https:\u002F\u002Fjessezhang92.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jian Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=6CIDtZQAAAAJ)\u003C\u002Fli> \u003Cli>[Zhenheng Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Ds5wwRoAAAAJ)\u003C\u002Fli> \u003Cli>[Ying Tai](https:\u002F\u002Ftyshiwo.github.io\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNJU-PCALab\u002FSTAR?style=social)](https:\u002F\u002Fgithub.com\u002FNJU-PCALab\u002FSTAR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2501.02976)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fali-vilab\u002FVGen), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FVchitect\u002FVEnhancer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNJU-PCALab\u002FOpenVid-1M), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhpcaitech\u002FOpen-Sora\u002Ftree\u002Fmain\u002Ftools\u002Fcaption#pllava-captioning)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FSherryX\u002FSTAR)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-star-spatial-temporal-augmentation-with-text-to-video-models-for-real-world-video-ff0ddcc6352f)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnju-pcalab.github.io\u002Fprojects\u002FSTAR)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fhx0zrql-SrU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1K8A1U_BNpAteRhhW9A8pAYs6LWjItQs_) | 22.01.2025 |\n| InvSR | Image super-resolution technique based on diffusion inversion, aiming at harnessing the rich image priors encapsulated in large pre-trained diffusion models to improve SR performance | \u003Cul>\u003Cli>[Zongsheng Yue](https:\u002F\u002Fzsyoaoa.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kang Liao](https:\u002F\u002Fkangliao929.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FzsyOAOA\u002FInvSR?style=social)](https:\u002F\u002Fgithub.com\u002FzsyOAOA\u002FInvSR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.09013)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcsjcai\u002FRealSR)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FOAOA\u002FInvSR), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fstabilityai\u002Fsd-turbo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FmCd7ODSmols)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1hjgCFnAU4oUUhh9VRfTwsFN1AiIjdcSR) | 21.01.2025 |\n| ModernBERT | Bringing modern model optimizations to encoder-only models and representing a major Pareto improvement over older encoders | \u003Cul>\u003Cli>[Benjamin Warner](https:\u002F\u002Fbenjaminwarner.dev\u002F)\u003C\u002Fli> \u003Cli>[Antoine Chaffin](https:\u002F\u002Fantoine.chaffin.fr\u002F)\u003C\u002Fli> \u003Cli>[Benjamin Clavié](https:\u002F\u002Fben.clavie.eu\u002F)\u003C\u002Fli> \u003Cli>[Orion Weller](https:\u002F\u002Forionweller.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Oskar Hallström](https:\u002F\u002Fgithub.com\u002Fohallstrom)\u003C\u002Fli> \u003Cli>[Said Taghadouini](https:\u002F\u002Fgithub.com\u002Fstaghado)\u003C\u002Fli> \u003Cli>[Alexis Gallagher](https:\u002F\u002Falexisgallagher.com\u002F)\u003C\u002Fli> \u003Cli>[Raja Biswas](https:\u002F\u002Fgithub.com\u002Frbiswasfc)\u003C\u002Fli> \u003Cli>[Faisal Ladhak](https:\u002F\u002Fgithub.com\u002Ffladhak)\u003C\u002Fli> \u003Cli>[Tom Aarsen](https:\u002F\u002Fwww.tomaarsen.com\u002Fhome)\u003C\u002Fli> \u003Cli>[Nathan Cooper](https:\u002F\u002Fnathancooper.io\u002F)\u003C\u002Fli> \u003Cli>[Griffin Adams](https:\u002F\u002Fgithub.com\u002Fgriff4692)\u003C\u002Fli> \u003Cli>[Jeremy Howard](https:\u002F\u002Fjeremy.fast.ai\u002F)\u003C\u002Fli> \u003Cli>[Iacopo Poli](https:\u002F\u002Fgithub.com\u002Fiacolippo)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAnswerDotAI\u002FModernBERT?style=social)](https:\u002F\u002Fgithub.com\u002FAnswerDotAI\u002FModernBERT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2412.13663)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmosaicml\u002Fexamples\u002Ftree\u002Fmain\u002Fexamples\u002Fbenchmarks\u002Fbert), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flightonai\u002Fpylate)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fmodernbert)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@sahin.samia\u002Fmodernbert-the-next-generation-of-encoder-models-a-guide-to-using-and-fine-tuning-for-nlp-tasks-995a50d5232f)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FZ1Dl3juwtSU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-kXUWeNcfUw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FZWo6Q8580sA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F92HKsDHD9XI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAnswerDotAI\u002FModernBERT\u002Fblob\u002Fmaster\u002Fexamples\u002Ffinetune_modernbert_on_glue.ipynb) | 22.12.2024 |\n| GraphCast | Learning skillful medium-range global weather forecasting | \u003Cul>\u003Cli>[Rémi Lam](https:\u002F\u002Fgithub.com\u002Fremilam)\u003C\u002Fli> \u003Cli>[Alvaro Sanchez-Gonzalez](https:\u002F\u002Fgithub.com\u002Falvarosg)\u003C\u002Fli> \u003Cli>[Matthew Willson](https:\u002F\u002Fgithub.com\u002Fmjwillson)\u003C\u002Fli> \u003Cli>[Peter Wirnsberger](https:\u002F\u002Fpewi.org\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Meire Fortunato](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=_fMHSIUAAAAJ)\u003C\u002Fli> \u003Cli>[Ferran Alet](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=1lmBq3QAAAAJ)\u003C\u002Fli> \u003Cli>[Suman Ravuri](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsuman-ravuri-81928082)\u003C\u002Fli> \u003Cli>[Timo Ewalds](https:\u002F\u002Fgithub.com\u002Ftewalds)\u003C\u002Fli> \u003Cli>[Zach Eaton-Rosen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=mQ3zD_wAAAAJ)\u003C\u002Fli> \u003Cli>[Weihua Hu](https:\u002F\u002Fweihua916.github.io\u002F)\u003C\u002Fli> \u003Cli>[Alexander Merose](https:\u002F\u002Falex.merose.com\u002F)\u003C\u002Fli> \u003Cli>[Stephan Hoyer](https:\u002F\u002Fstephanhoyer.com\u002F)\u003C\u002Fli> \u003Cli>[George Holland](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fg-aracil-holland)\u003C\u002Fli> \u003Cli>[Oriol Vinyals](https:\u002F\u002Fresearch.google\u002Fpeople\u002Foriol-vinyals\u002F)\u003C\u002Fli> \u003Cli>[Jacklynn Stott](https:\u002F\u002Flinkedin.com\u002Fin\u002Fjacklynnstott)\u003C\u002Fli> \u003Cli>[Alexander Pritzel](https:\u002F\u002Fgithub.com\u002Fa-pritzel)\u003C\u002Fli> \u003Cli>[Shakir Mohamed](https:\u002F\u002Fwww.shakirm.com\u002F)\u003C\u002Fli> \u003Cli>[Peter Battaglia](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=nQ7Ij30AAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_95a3b8c45abb.png)](https:\u002F\u002Fdoi.org\u002F10.1126\u002Fscience.adi2336) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Fgraphcast?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fgraphcast) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.12794)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fwww.ecmwf.int\u002Fen\u002Fforecasts\u002Fdatasets\u002Freanalysis-datasets\u002Fera5)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fdiscover\u002Fblog\u002Fgraphcast-ai-model-for-faster-and-more-accurate-global-weather-forecasting\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fchex), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdask\u002Fdask), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fjaxline), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Ftree), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmikedh\u002Ftrimesh)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Ftowardsdatascience.com\u002Fgraphcast-how-to-get-things-done-f2fd5630c5fb)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBufUW7h9TB8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPD1v5PCJs_o), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FEul-JN9Nwb0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBTyhgp9Hugc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FaJ_H4exg0xU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Fgraphcast\u002Fblob\u002Fmaster\u002Fgraphcast_demo.ipynb) | 04.12.2024 |\n| TAPIR | Tracking Any Point with per-frame Initialization and temporal Refinement | \u003Cul>\u003Cli>[Carl Doersch](http:\u002F\u002Fwww.carldoersch.com\u002F)\u003C\u002Fli> \u003Cli>[Yi Yang](https:\u002F\u002Fyangyi02.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mel Vecerik](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Jvi_XPAAAAAJ)\u003C\u002Fli> \u003Cli>[Dilara Gokay](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=cnbENAEAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ankush Gupta](https:\u002F\u002Fankushgupta.org\u002F)\u003C\u002Fli> \u003Cli>[Yusuf Aytar](https:\u002F\u002Fpeople.csail.mit.edu\u002Fyusuf\u002F)\u003C\u002Fli> \u003Cli>[Joao Carreira](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=IUZ-7_cAAAAJ)\u003C\u002Fli> \u003Cli>[Andrew Zisserman](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~az\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Ftapnet?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Ftapnet) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.08637), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.15975)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fdeepmind-tapir.github.io\u002F), [blog post](https:\u002F\u002Fdeepmind-tapir.github.io\u002Fblogpost.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fwww.deepmind.com\u002Fopen-source\u002Fkinetics)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fkubric\u002Ftree\u002Fmain\u002Fchallenges\u002Fpoint_tracking)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@jumabek4044\u002Fwhat-is-tapir-tracking-any-point-with-per-frame-initialization-and-temporal-refinement-and-how-it-bdad9946dc53)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002F58168e8a92994655d6da3939e7cc0918-Abstract-Datasets_and_Benchmarks.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2HSHofqoJ9M), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FI1DQJH3v7Nk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Ftapnet\u002Fblob\u002Fmaster\u002Fcolabs\u002Fcausal_tapir_demo.ipynb) | 30.11.2024 |\n| ConsisID | Tuning-free DiT-based controllable IPT2V model to keep human identity consistent in the generated video | \u003Cul>\u003Cli>[Shenghai Yuan](https:\u002F\u002Fshyuanbest.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jinfa Huang](https:\u002F\u002Finfaaa.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xianyi He](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=f72qwN8AAAAJ)\u003C\u002Fli> \u003Cli>[Yunyuan Ge](https:\u002F\u002Fgithub.com\u002Fyunyangge)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yujun Shi](https:\u002F\u002Fyujun-shi.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liuhan Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=eALObLQAAAAJ)\u003C\u002Fli> \u003Cli>[Jiebo Luo](https:\u002F\u002Fwww.cs.rochester.edu\u002Fu\u002Fjluo\u002F)\u003C\u002Fli> \u003Cli>[Li Yuan](https:\u002F\u002Fyuanli2333.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPKU-YuanGroup\u002FConsisID?style=social)](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FConsisID) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2411.17440)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FXueZeyue\u002FDanceGRPO), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FPKU-YuanGroup\u002FOpenS2V-Nexus), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fali-vilab\u002FTeaCache), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxdit-project\u002FxDiT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkijai\u002FComfyUI-CogVideoXWrapper), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffeizc\u002FIngredients), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Faigc-apps\u002FEasyAnimate), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Faigc-apps\u002FVideoX-Fun), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjunjiehe96\u002FUniPortrait)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FBestWishYsh\u002FConsisID-preview-Data), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FBestWishYsh\u002FConsisID-preview-Space)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpku-yuangroup.github.io\u002FConsisID\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FAdinaYakup\u002Fstatus\u002F1862604191631573122), [\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002Fcamenduru\u002Fstatus\u002F1861957812152078701)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPhlgC-bI5SQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FConsisID-jupyter\u002Fblob\u002Fmain\u002FConsisID_jupyter.ipynb) | 28.11.2024 |\n| T2M-GPT | Conditional generative framework based on Vector Quantised-Variational AutoEncoder and Generative Pre-trained Transformer for human motion generation from textural descriptions | \u003Cul>\u003Cli>[Jianrong Zhang](https:\u002F\u002Fgithub.com\u002FJiro-zhang)\u003C\u002Fli> \u003Cli>[Yangsong Zhang](https:\u002F\u002Fgithub.com\u002FMael-zys)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002Facademic\u002F)\u003C\u002Fli> \u003Cli>[Shaoli Huang](https:\u002F\u002Fshaoli-huang.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yong Zhang](https:\u002F\u002Fyzhang2016.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hongwei Zhao](https:\u002F\u002Fteachers.jlu.edu.cn\u002Fzhaohongwei\u002Fen\u002Findex.htm)\u003C\u002Fli> \u003Cli>[Hongtao Lu](https:\u002F\u002Fwww.cs.sjtu.edu.cn\u002Fen\u002FPeopleDetail.aspx?id=156)\u003C\u002Fli> \u003Cli>[Xi Shen](https:\u002F\u002Fxishen0220.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c2afe1505606.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52729.2023.01415) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FMael-zys\u002FT2M-GPT?style=social)](https:\u002F\u002Fgithub.com\u002FMael-zys\u002FT2M-GPT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.06052)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEricGuo5513\u002FHumanML3D), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEricGuo5513\u002Ftext-to-motion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FGuyTevet\u002Fmotion-diffusion-model), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEricGuo5513\u002FTM2T)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fvumichien\u002FT2M-GPT), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fvumichien\u002Fgenerate_human_motion)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@kaveh.kamali\u002Ft2m-gpt-pioneering-human-motion-generation-from-textual-descriptions-48dc62b5cd7a)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmael-zys.github.io\u002FT2M-GPT\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F09K2cx9P0_0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Vy69w2q2d-Hg19F-KibqG0FRdpSj3L4O) | 24.11.2024 |\n| PuLID | Pure and Lightning ID customization, a tuning-free ID customization method for text-to-image generation | \u003Cul>\u003Cli>[Zinan Guo](https:\u002F\u002Fgithub.com\u002Fguozinan126)\u003C\u002Fli> \u003Cli>[Yanze Wu](https:\u002F\u002Ftothebeginning.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhuowei Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ow1jGJkAAAAJ)\u003C\u002Fli> \u003Cli>[Lang Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=h5xex20AAAAJ)\u003C\u002Fli> \u003Cli>[Qian He](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=9rWWCgUAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FToTheBeginning\u002FPuLID?style=social)](https:\u002F\u002Fgithub.com\u002FToTheBeginning\u002FPuLID) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.16022)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcubiq\u002FPuLID_ComfyUI), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZHO-ZHO-ZHO\u002FComfyUI-PuLID-ZHO), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMikubill\u002Fsd-webui-controlnet\u002Fpull\u002F2838)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fcomfyui\u002Fcomments\u002F1cnv269\u002Fpulid_pure_and_lightning_id_customization_via\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FPuLID-jupyter\u002Fblob\u002Fmain\u002FPuLID_jupyter.ipynb) | 09.11.2024 |\n| CoTracker | Architecture that jointly tracks multiple points throughout an entire video | \u003Cul>\u003Cli>[Nikita Karaev](https:\u002F\u002Fnikitakaraevv.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ignacio Rocco](https:\u002F\u002Fwww.irocco.info\u002F)\u003C\u002Fli> \u003Cli>[Benjamin Graham](https:\u002F\u002Fai.meta.com\u002Fpeople\u002Fbenjamin-graham\u002F)\u003C\u002Fli> \u003Cli>[Natalia Neverova](https:\u002F\u002Fnneverova.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Andrea Vedaldi](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vedaldi\u002F)\u003C\u002Fli> \u003Cli>[Christian Rupprecht](https:\u002F\u002Fchrirupp.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fco-tracker?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fco-tracker) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.07635), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.11898)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbenjiebob\u002FBADJA)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fco-tracker.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fw5QVc7BVGPA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fco-tracker\u002Fblob\u002Fmain\u002Fnotebooks\u002Fdemo.ipynb) | 16.10.2024 |\n| PIFu | Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization | \u003Cul>\u003Cli>[Ryota Natsume](https:\u002F\u002Fgithub.com\u002Fnanopoteto)\u003C\u002Fli> \u003Cli>[Shunsuke Saito](https:\u002F\u002Fshunsukesaito.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zeng Huang](https:\u002F\u002Fzeng.science\u002F)\u003C\u002Fli> \u003Cli>[Angjoo Kanazawa](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~kanazawa\u002F)\u003C\u002Fli> \u003Cli>[Hao Li](http:\u002F\u002Fhao.li)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_d66c3b54eb70.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV.2019.00239) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fshunsukesaito\u002FPIFu?style=social)](https:\u002F\u002Fgithub.com\u002Fshunsukesaito\u002FPIFu) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.05172)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=S1FpjwKqtPs)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1GFSsqP2BWz4gtq0e-nki00ZHSirXwFyY) | 08.10.2024 |\n| DifFace | Method that is capable of coping with unseen and complex degradations more gracefully without complicated loss designs | \u003Cul>\u003Cli>[Zongsheng Yue](https:\u002F\u002Fzsyoaoa.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_28a1a4fb4867.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTPAMI.2024.3432651) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FzsyOAOA\u002FDifFace?style=social)](https:\u002F\u002Fgithub.com\u002FzsyOAOA\u002FDifFace) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.06512)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fimproved-diffusion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepcam-cn\u002Fyolov5-face), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002Ffacexlib)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FOAOA\u002FDifFace)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1BNtoPPRuJwNDvqfwDOOmD9XJyF05Zh4m) | 05.10.2024 |\n| Segment Anything 2 | Foundation model towards solving promptable visual segmentation in images and videos | \u003Cul>\u003Cli>[Nikhila Ravi](https:\u002F\u002Fnikhilaravi.com\u002F)\u003C\u002Fli> \u003Cli>[Valentin Gabeur](https:\u002F\u002Fgabeur.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuan-Ting Hu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=E8DVVYQAAAAJ)\u003C\u002Fli> \u003Cli>[Ronghang Hu](https:\u002F\u002Fronghanghu.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chaitanya Ryali](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4LWx24UAAAAJ)\u003C\u002Fli> \u003Cli>[Tengyu Ma](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=VeTSl0wAAAAJ)\u003C\u002Fli> \u003Cli>[Haitham Khedr](https:\u002F\u002Fhkhedr.com\u002F)\u003C\u002Fli> \u003Cli>[Roman Rädle](https:\u002F\u002Fscholar.google.de\u002Fcitations?user=Tpt57v0AAAAJ)\u003C\u002Fli> \u003Cli>[Chloé Rolland](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=n-SnMhoAAAAJ)\u003C\u002Fli> \u003Cli>[Laura Gustafson](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=c8IpF9gAAAAJ)\u003C\u002Fli> \u003Cli>[Eric Mintun](https:\u002F\u002Fericmintun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Junting Pan](https:\u002F\u002Fjunting.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kalyan Vasudev](lwala](https:\u002F\u002Fscholar.google.co.in\u002Fcitations?user=m34oaWEAAAAJ)\u003C\u002Fli> \u003Cli>[Nicolas Carion](https:\u002F\u002Fwww.nicolascarion.com\u002F)\u003C\u002Fli> \u003Cli>[Chao-Yuan](u](https:\u002F\u002Fchaoyuan.org\u002F)\u003C\u002Fli> \u003Cli>[Ross Girshick](https:\u002F\u002Fwww.rossgirshick.info\u002F)\u003C\u002Fli> \u003Cli>[Piotr Dollár](https:\u002F\u002Fpdollar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Christoph Feichtenhofer](https:\u002F\u002Ffeichtenhofer.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fsegment-anything-2?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fsegment-anything-2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.00714)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fsam2.metademolab.com\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzsef123\u002FConnected_components_PyTorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fmodels?search=facebook\u002Fsam2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fresearch\u002Fpublications\u002Fsam-2-segment-anything-in-images-and-videos\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fdatasets\u002Fsegment-anything-video), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.meta.com\u002Fblog\u002Fsegment-anything-2)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fai.meta.com\u002Fsam2\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FAIatMeta\u002Fstatus\u002F1818055906179105010)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=w-cmMcMZoZ4&t=2325s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FO8QdvZbRDp4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FDv003fTyO-Y), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIW7jFq3vQbw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fsegment-anything-2\u002Fblob\u002Fmain\u002Fnotebooks\u002Fimage_predictor_example.ipynb) | 01.10.2024 |\n| Open-Unmix | A deep neural network reference implementation for music source separation, applicable for researchers, audio engineers and artists | \u003Cul>\u003Cli>[Fabian-Robert Stöter](http:\u002F\u002Ffaroit.com\u002F)\u003C\u002Fli> \u003Cli>[Antoine Liutkus](https:\u002F\u002Fgithub.com\u002Faliutkus)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_7aec23c2275d.png)](https:\u002F\u002Fdoi.org\u002F10.21105\u002Fjoss.01667) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsigsep\u002Fopen-unmix-pytorch?style=social)](https:\u002F\u002Fgithub.com\u002Fsigsep\u002Fopen-unmix-pytorch) \u003Cul>\u003Cli>[data](https:\u002F\u002Fsigsep.github.io\u002Fdatasets\u002Fmusdb.html#musdb18-compressed-stems)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsigsep\u002Fnorbert)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsigsep.github.io\u002Fopen-unmix\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fmusic-source-separation-on-musdb18?p=open-unmix-a-reference-implementation-for)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLhA3b2k8R3t0VpYCpCTU2B1h604rvnV4N)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1mijF0zGWxN-KaxTnd0q6hayAlrID5fEQ) | 25.09.2024 |\n| Deep Painterly Harmonization | Algorithm produces significantly better results than photo compositing or global stylization techniques and that it enables creative painterly edits that would be otherwise difficult to achieve | \u003Cul>\u003Cli>[Fujun Luan](https:\u002F\u002Fluanfujun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sylvain Paris](http:\u002F\u002Fpeople.csail.mit.edu\u002Fsparis\u002F)\u003C\u002Fli> \u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli> \u003Cli>[Kavita Bala](https:\u002F\u002Fwww.cs.cornell.edu\u002F~kb\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fluanfujun\u002Fdeep-painterly-harmonization?style=social)](https:\u002F\u002Fgithub.com\u002Fluanfujun\u002Fdeep-painterly-harmonization) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.03189), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1701.08893)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjcjohnson\u002Fneural-style), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftorch\u002Ftorch7), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fszagoruyko\u002Floadcaffe)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgist\u002Feyaler\u002F5303782669fb43510d398bd346c6e3e6\u002Fdeep-painterly-harmonization.ipynb) | 23.09.2024 |\n| CogVideo | Large-scale text-to-video generation model based on diffusion transformer, which can generate 10-second continuous videos aligned with text prompt, with a frame rate of 16 fps and resolution of 768 * 1360 pixels | \u003Cul>\u003Cli>[Zhuoyi Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=tgAt-gEAAAAJ)\u003C\u002Fli> \u003Cli>[Jiayan Teng](https:\u002F\u002Ftengjiayan20.github.io\u002F)\u003C\u002Fli> \u003Cli>[Wendi Zheng](https:\u002F\u002Fgithub.com\u002Fminkowski0125)\u003C\u002Fli> \u003Cli>[Ming Ding](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Va50YzkAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Shiyu Huang](https:\u002F\u002Fgithub.com\u002Fhuangshiyu13)\u003C\u002Fli> \u003Cli>[Jiazheng Xu](https:\u002F\u002Fgithub.com\u002Fxujz18)\u003C\u002Fli> \u003Cli>[Yuanming Yang](https:\u002F\u002Fgithub.com\u002Fyomi1117)\u003C\u002Fli> \u003Cli>[Wenyi Hong](https:\u002F\u002Fgithub.com\u002Fwenyihong)\u003C\u002Fli> \u003Cli>[Xiaohan Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RKyE8o0AAAAJ)\u003C\u002Fli> \u003Cli>[Guanyu Feng](https:\u002F\u002Fgithub.com\u002Fjiguanglizipao)\u003C\u002Fli> \u003Cli>[Da Yin](https:\u002F\u002Fsomefive.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuxuan Zhang](https:\u002F\u002Fgithub.com\u002FzRzRzRzRzRzRzR)\u003C\u002Fli> \u003Cli>[Weihan Wang](https:\u002F\u002Fgithub.com\u002Fmactavish91)\u003C\u002Fli> \u003Cli>[Yean Cheng](https:\u002F\u002Fliquidammonia.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bin Xu](https:\u002F\u002Fkeg.cs.tsinghua.edu.cn\u002Fpersons\u002Fxubin\u002F)\u003C\u002Fli> \u003Cli>[Xiaotao Gu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=YR4Lp0QAAAAJ)\u003C\u002Fli> \u003Cli>[Yuxiao Dong](https:\u002F\u002Fkeg.cs.tsinghua.edu.cn\u002Fyuxiao\u002F)\u003C\u002Fli> \u003Cli>[Jie Tang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=XfFqozoAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTHUDM\u002FCogVideo?style=social)](https:\u002F\u002Fgithub.com\u002FTHUDM\u002FCogVideo) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2408.06072), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.15868)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fyzy-thu.github.io\u002FCogVideoX-demo\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FdCGfUsagrD)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTHUDM\u002FCogKit), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fa-r-r-o-w\u002Fcogvideox-factory), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsayakpaul\u002Fdiffusers-torchao), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTHUDM\u002FSwissArmyTransformer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fpytorch\u002Fao), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Foptimum-quanto), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Faigc-apps\u002FCogVideoX-Fun), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fthu-ml\u002FRIFLEx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fpinokiofactory\u002Fcogstudio), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkijai\u002FComfyUI-CogVideoXWrapper), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNUS-HPC-AI-Lab\u002FVideoSys), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffeizc\u002FCogvideX-Interpolation), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTheDenk\u002Fcogvideox-controlnet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FVideoVerses\u002FVideoTuna), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxdit-project\u002FxDiT)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FTHUDM\u002FCogVideoX-5B), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FTHUDM\u002FCogVideoX1.5-5B-SAT), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FTHUDM\u002FCogVideoX-5B-Space), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FTHUDM\u002Fcogvlm2-llama3-caption)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1el6uy0\u002Fcogvideox_texttovideo_diffusion_models_with_an\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5UCkMzP2VLE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUD3ZFLj-3uE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FChS4JGijwPk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPPGYNmrVG58), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3IRnXpl9Zmo)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1pCe5s0bC_xuXbBlpvIH1z0kfdTLQPzCS) | 18.09.2024 |\n| audio2photoreal | Framework for generating full-bodied photorealistic avatars that gesture according to the conversational dynamics of a dyadic interaction | \u003Cul>\u003Cli>[Evonne Ng](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~evonne_ng\u002F)\u003C\u002Fli> \u003Cli>[Javier Romero](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Wx62iOsAAAAJ)\u003C\u002Fli> \u003Cli>[Timur Bagautdinov](https:\u002F\u002Fscholar.google.ch\u002Fcitations?user=oLi7xJ0AAAAJ)\u003C\u002Fli> \u003Cli>[Shaojie Bai](https:\u002F\u002Fjerrybai1995.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Trevor Darrell](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~trevor\u002F)\u003C\u002Fli> \u003Cli>[Angjoo Kanazawa](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~kanazawa\u002F)\u003C\u002Fli> \u003Cli>[Alexander Richard](https:\u002F\u002Falexanderrichard.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Faudio2photoreal?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Faudio2photoreal) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.01885)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fca_body)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~evonne_ng\u002Fprojects\u002Faudio2photoreal\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FY0GMaMtUynQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1A6EwKM3PeX7dcKV66zxQWuP-v_dKlX_0) | 13.09.2024 |\n| Fast Segment Anything | CNN Segment Anything Model trained using only 2% of the SA-1B dataset published by SAM authors | \u003Cul>\u003Cli>[Xu Zhao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=F0cYEyAAAAAJ)\u003C\u002Fli> \u003Cli>[Wenchao Ding](https:\u002F\u002Fgithub.com\u002Fberry-ding)\u003C\u002Fli> \u003Cli>[Yongqi An](https:\u002F\u002Fgithub.com\u002Fan-yongqi)\u003C\u002Fli> \u003Cli>[Yinglong Du](https:\u002F\u002Fgithub.com\u002FYinglongDu)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Tao Yu](https:\u002F\u002Fgithub.com\u002Ftianjinren)\u003C\u002Fli> \u003Cli>[Min Li](https:\u002F\u002Fgithub.com\u002Flimin2021)\u003C\u002Fli> \u003Cli>[Ming Tang](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FMing-Tang-2)\u003C\u002Fli> \u003Cli>[Jinqiao Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=7_BkyxEAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCASIA-IVA-Lab\u002FFastSAM?style=social)](https:\u002F\u002Fgithub.com\u002FCASIA-IVA-Lab\u002FFastSAM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.12156), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10003)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FChuRuaNh0\u002FFastSam_Awsome_TensorRT)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@mahimairaja\u002Fso-what-exactly-is-fastsam-the-ultimate-guide-ddae21d3b486)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyHNPyqazYYU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FSslzS0AsiAw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FqvqkjP1wCDE)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1oX14f6IneGGw612WgVlAiy91UHwFAvr9) | 10.09.2024 |\n| Neuralangelo | Framework for high-fidelity 3D surface reconstruction from RGB video captures | \u003Cul>\u003Cli>[Zhaoshuo Li](https:\u002F\u002Fmli0603.github.io\u002F)\u003C\u002Fli> \u003Cli>[Thomas Müller](https:\u002F\u002Ftom94.net\u002F)\u003C\u002Fli> \u003Cli>[Alex Evans](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ToqGImkAAAAJ)\u003C\u002Fli> \u003Cli>[Russell Taylor](https:\u002F\u002Fwww.cs.jhu.edu\u002F~rht\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Mathias Unberath](https:\u002F\u002Fmathiasunberath.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ming-Yu Liu](https:\u002F\u002Fmingyuliu.net\u002F)\u003C\u002Fli> \u003Cli>[Chen-Hsuan Lin](https:\u002F\u002Fchenhsuanlin.bitbucket.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNVlabs\u002Fneuralangelo?style=social)](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fneuralangelo) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.03092)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fblogs.nvidia.com\u002Fblog\u002F2023\u002F06\u002F01\u002Fneuralangelo-ai-research-3d-reconstruction\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmli0603\u002FBlenderNeuralangelo)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fresearch.nvidia.com\u002Flabs\u002Fdir\u002Fneuralangelo\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPQMNCXR-WF8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQpdw3SW54kI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FlC2uPDfaTcE)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1i16s8W_OV0Hd3-PIuo64JKDwwdOesgXQ) | 02.09.2024 |\n| SPIN | Learning to Reconstruct 3D Human Pose and Shape via Model-fitting in the Loop | \u003Cul>\u003Cli>[Nikos Kolotouros](https:\u002F\u002Fwww.nikoskolot.com\u002F)\u003C\u002Fli> \u003Cli>[Georgios Pavlakos](https:\u002F\u002Fgeopavlakos.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli> \u003Cli>[Kostas Daniilidis](https:\u002F\u002Fwww.cis.upenn.edu\u002F~kostas\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_aed0085d183c.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV.2019.00234) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fnkolot\u002FSPIN?style=social)](https:\u002F\u002Fgithub.com\u002Fnkolot\u002FSPIN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.12828)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fchaneyk\u002Fspin)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplify-x), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCMU-Perceptual-Computing-Lab\u002Fopenpose)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.nikoskolot.com\u002Fprojects\u002Fspin\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1uH2JtavOtDrFl6RsipyIncCSr19GWW4x) | 21.08.2024 |\n| YOLOv10 | Aim to further advance the performance-efficiency boundary of YOLOs from both the post-processing and model architecture | \u003Cul>\u003Cli>[Ao Wang](https:\u002F\u002Fgithub.com\u002Fjameslahm)\u003C\u002Fli> \u003Cli>[Hui Chen](https:\u002F\u002Fhuichen24.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kai Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=bZQX708AAAAJ)\u003C\u002Fli> \u003Cli>[Zijia Lin](https:\u002F\u002Fsites.google.com\u002Fsite\u002Flinzijia72)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jungong Han](https:\u002F\u002Fjungonghan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Guiguang Ding](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=B7F3yt4AAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTHU-MIG\u002Fyolov10?style=social)](https:\u002F\u002Fgithub.com\u002FTHU-MIG\u002Fyolov10) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.14458)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Flearnopencv.com\u002Fyolov10\u002F)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fopenbayes.com\u002Fconsole\u002Fpublic\u002Ftutorials\u002Fim29uYrnIoz)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frlggyp\u002FYOLOv10-OpenVINO-CPP-Inference), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FSeeed-Projects\u002Fjetson-examples\u002Fblob\u002Fmain\u002FreComputer\u002Fscripts\u002Fyolov10\u002FREADME.md), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkaylorchen\u002Frk3588-yolo-demo), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenvinotoolkit\u002Fopenvino_notebooks\u002Fblob\u002Flatest\u002Fnotebooks\u002Fyolov10-optimization\u002Fyolov10-optimization.ipynb), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsujanshresstha\u002FYOLOv10_DeepSORT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCVHub520\u002FX-AnyLabeling), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDanielSarmiento04\u002Fyolov10cpp), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flyuwenyu\u002FRT-DETR)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcollections\u002Fjameslahm\u002Fyolov10-665b0d90b0b5bb85129460c2), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fjameslahm\u002FYOLOv10), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fkadirnar\u002FYolov10), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FXenova\u002Fyolov10-web)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@batuhansenerr\u002Fyolov10-custom-object-detection-bd7298ddbfd3), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@sunidhi.ashtekar\u002Fyolov10-revolutionizing-real-time-object-detection-72ef04ad441a)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FGPTFutureScience\u002Fcomments\u002F1d34rj1\u002Fyolov10_the_future_of_realtime_object_detection\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F29tnSxhB3CY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2ZFJbeJXXDM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FwM6nO75keOQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Froboflow-ai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnotebooks\u002Ftrain-yolov10-object-detection-on-custom-dataset.ipynb) | 20.08.2024 |\n| SpecVQGAN | Taming the visually guided sound generation by shrinking a training dataset to a set of representative vectors | \u003Cul>\u003Cli>[Vladimir Iashin](https:\u002F\u002Fiashin.ai\u002F)\u003C\u002Fli> \u003Cli>[Esa Rahtu](https:\u002F\u002Fesa.rahtu.fi\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fv-iashin\u002FSpecVQGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fv-iashin\u002FSpecVQGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F2110.08791), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.09841), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.00937), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.00820), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1712.01393), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1512.08512)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FPeihaoChen\u002Fregnet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftoshas\u002Ftorch-fidelity), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdescriptinc\u002Fmelgan-neurips), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Flyra)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fiashin.ai\u002FSpecVQGAN)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FFoley_(filmmaking)), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FRow-_and_column-major_order), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FKullback%E2%80%93Leibler_divergence)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=Bucb3nAa398)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1pxTIMweAKApJZ3ZFqyBee3HtMqFpnwQ0) | 12.07.2024 |\n| LivePortrait | Video-driven portrait animation framework with a focus on better generalization, controllability, and efficiency for practical usage | \u003Cul>\u003Cli>[Jianzhu Guo](https:\u002F\u002Fguojianzhu.com\u002F)\u003C\u002Fli> \u003Cli>[Dingyun Zhang](https:\u002F\u002Fgithub.com\u002FDingyunZhang)\u003C\u002Fli> \u003Cli>[Xiaoqiang Liu](https:\u002F\u002Fgithub.com\u002FLiu-lxq)\u003C\u002Fli> \u003Cli>[Zhizhou Zhong](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=t88nyvsAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yuan Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=_8k1ubAAAAAJ)\u003C\u002Fli> \u003Cli>[Pengfei Wan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=P6MraaYAAAAJ)\u003C\u002Fli> \u003Cli>[Di Zhang](https:\u002F\u002Fopenreview.net\u002Fprofile?id=~Di_ZHANG3)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FKwaiVGI\u002FLivePortrait?style=social)](https:\u002F\u002Fgithub.com\u002FKwaiVGI\u002FLivePortrait) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.03168)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkijai\u002FComfyUI-LivePortraitKJ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fshadowcz007\u002Fcomfyui-liveportrait), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhanglonghao1992\u002FOne-Shot_Free-View_Neural_Talking_Head_Synthesis), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002FSPADE), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FKwaiVGI\u002FLivePortrait)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fliveportrait.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1dvepjx\u002Fliveportrait_efficient_portrait_animation_with\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuyjSTAOY7yI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8-IcDDmiUMM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FaFcS31OWMjE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbRHf2oQwgG4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FFPtpNrmuwXk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FwG7oPp01COg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FLivePortrait-jupyter\u002Fblob\u002Fmain\u002FLivePortrait_jupyter.ipynb) | 10.07.2024 |\n| Wav2Lip | A Lip Sync Expert Is All You Need for Speech to Lip Generation In the Wild | \u003Cul>\u003Cli>[Prajwal Renukanand](https:\u002F\u002Fgithub.com\u002Fprajwalkr)\u003C\u002Fli> \u003Cli>[Rudrabha Mukhopadhyay](https:\u002F\u002Frudrabha.github.io\u002F)\u003C\u002Fli> \u003Cli>[Vinay Namboodiri](https:\u002F\u002Fvinaypn.github.io\u002F)\u003C\u002Fli> \u003Cli>[C. V. Jawahar](https:\u002F\u002Ffaculty.iiit.ac.in\u002F~jawahar\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2f88fb4d252e.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3394171.3413532) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FRudrabha\u002FWav2Lip?style=social)](https:\u002F\u002Fgithub.com\u002FRudrabha\u002FWav2Lip) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.10010)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vgg\u002Fdata\u002Flip_reading\u002Flrs2.html)\u003C\u002Fli>\u003Cli>[demo](http:\u002F\u002Fbhaasha.iiit.ac.in\u002Flipsync\u002F)\u003C\u002Fli>\u003Cli>[project](http:\u002F\u002Fcvit.iiit.ac.in\u002Fresearch\u002Fprojects\u002Fcvit-projects\u002Fa-lip-sync-expert-is-all-you-need-for-speech-to-lip-generation-in-the-wild\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=0fXaDCZNOJc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Feyaler\u002Favatars4all\u002Fblob\u002Fmaster\u002Fmelaflefon.ipynb) | 27.06.2024 |\n| FELIX | Feature Engineering with LLMs for Interpretability and Explainability, a novel approach harnessing the vast world knowledge embedded in pre-trained Large Language Models to automatically generate a set of features describing the data | \u003Cul>\u003Cli>[Simon Malberg](https:\u002F\u002Fgithub.com\u002Fsimonmalberg)\u003C\u002Fli> \u003Cli>[Edoardo Mosca](https:\u002F\u002Fedoardomosca.github.io\u002F)\u003C\u002Fli> \u003Cli>[Georg Groh](https:\u002F\u002Fsocvm1.cit.tum.de\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_4c82bcaa094d.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-70359-1_14) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsimonmalberg\u002Ffelix?style=social)](https:\u002F\u002Fgithub.com\u002Fsimonmalberg\u002Ffelix)  | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsimonmalberg\u002Ffelix\u002Fblob\u002Fmain\u002FPaper\u002FFELIX.ipynb) | 13.06.2024 |\n| PoolFormer | MetaFormer Is Actually What You Need for Vision | \u003Cul>\u003Cli>[Weihao Yu](https:\u002F\u002Fwhyu.me\u002F)\u003C\u002Fli> \u003Cli>[Mi Luo](https:\u002F\u002Fluomi97.github.io\u002F)\u003C\u002Fli> \u003Cli>[Pan Zhou](https:\u002F\u002Fpanzhous.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chenyang Si](https:\u002F\u002Fgithub.com\u002FChenyangSi)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yichen Zhou](https:\u002F\u002Fdblp.org\u002Fpid\u002F55\u002F10422.html)\u003C\u002Fli> \u003Cli>[Xinchao Wang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fsitexinchaowang\u002F)\u003C\u002Fli> \u003Cli>[Jiashi Feng](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fjshfeng\u002F)\u003C\u002Fli> \u003Cli>[Shuicheng Yan](https:\u002F\u002Fyanshuicheng.ai\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fb8d53f94a8d.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01055) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsail-sg\u002Fpoolformer?style=social)](https:\u002F\u002Fgithub.com\u002Fsail-sg\u002Fpoolformer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.11418)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frwightman\u002Fpytorch-image-models), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffvcore), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fapex)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002Fpoolformer)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsail-sg\u002Fpoolformer\u002Fblob\u002Fmain\u002Fmisc\u002Fpoolformer_demo.ipynb) | 01.06.2024 |\n| StoryDiffusion | Way of self-attention calculation, termed Consistent Self-Attention, that significantly boosts the consistency between the generated images and augments prevalent pretrained diffusion-based text-to-image models in a zero-shot manner | \u003Cul>\u003Cli>[Yupeng Zhou](https:\u002F\u002Fmmcheng.net\u002Fzyp\u002F)\u003C\u002Fli> \u003Cli>[Daquan Zhou](https:\u002F\u002Fgithub.com\u002Fzhoudaquan)\u003C\u002Fli> \u003Cli>[Ming-Ming Cheng](https:\u002F\u002Fmmcheng.net\u002Fcmm\u002F)\u003C\u002Fli> \u003Cli>[Jiashi Feng](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fjshfeng\u002F?pli=1)\u003C\u002Fli> \u003Cli>[Qibin Hou](https:\u002F\u002Fhouqb.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FHVision-NKU\u002FStoryDiffusion?style=social)](https:\u002F\u002Fgithub.com\u002FHVision-NKU\u002FStoryDiffusion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2405.01434)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGeNyP4VY9rE?si=qW1jcW_GbKutmKQv)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fstorydiffusion.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStoryDiffusion\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjZWRENqCl6I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGeNyP4VY9rE)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FHVision-NKU\u002FStoryDiffusion\u002Fblob\u002Fmain\u002FComic_Generation.ipynb) | 04.05.2024 |\n| FILM | A frame interpolation algorithm that synthesizes multiple intermediate frames from two input images with large in-between motion | \u003Cul>\u003Cli>[Fitsum Reda](https:\u002F\u002Ffitsumreda.github.io\u002F)\u003C\u002Fli> \u003Cli>[Janne Kontkanen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=MnXc4JQAAAAJ)\u003C\u002Fli> \u003Cli>[Eric Tabellion](http:\u002F\u002Fwww.tabellion.org\u002Fet\u002F)\u003C\u002Fli> \u003Cli>[Deqing Sun](https:\u002F\u002Fdeqings.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Caroline Pantofaru](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vKAKE1gAAAAJ)\u003C\u002Fli> \u003Cli>[Brian Curless](https:\u002F\u002Fhomes.cs.washington.edu\u002F~curless\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_d06e71d41283.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20071-7_15) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fframe-interpolation?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fframe-interpolation) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2202.04901)\u003C\u002Fli>\u003Cli>[data](http:\u002F\u002Fdata.csail.mit.edu\u002Ftofu\u002Ftestset\u002Fvimeo_interp_test.zip), [data](https:\u002F\u002Fvision.middlebury.edu\u002Fflow\u002Fdata), [data](https:\u002F\u002Fpeople.cs.umass.edu\u002F~hzjiang\u002Fprojects\u002Fsuperslomo\u002FUCF101_results.zip)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsniklaus\u002Fsoftmax-splatting\u002Fblob\u002Fmaster\u002Fbenchmark.py)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ffilm-net.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Ftutorials\u002Fload_data\u002Ftfrecord), [\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fapi_docs\u002Fpython\u002Ftf\u002Ftrain\u002FExample), [\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fguide\u002Fsaved_model)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FOAD-BieIjH4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1sK0uc-GJxmdnaxHhYqD2afRknakpdTNZ) | 03.05.2024 |\n| VoiceCraft | token infilling neural codec language model, that achieves state-of-the-art performance on both speech editing and zero-shot text-to-speech on audiobooks, internet videos, and podcasts | \u003Cul>\u003Cli>[Puyuan Peng](https:\u002F\u002Fjasonppy.github.io\u002F)\u003C\u002Fli> \u003Cli>[Po-Yao Huang](https:\u002F\u002Fberniebear.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shang-Wen Li](https:\u002F\u002Fswdanielli.github.io\u002F)\u003C\u002Fli> \u003Cli>[Abdelrahman Mohamed](https:\u002F\u002Fwww.cs.toronto.edu\u002F~asamir\u002F)\u003C\u002Fli> \u003Cli>[David Harwath](https:\u002F\u002Fwww.cs.utexas.edu\u002F~harwath\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjasonppy\u002FVoiceCraft?style=social)](https:\u002F\u002Fgithub.com\u002Fjasonppy\u002FVoiceCraft) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.16973)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flifeiteng\u002Fvall-e)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fpyp1\u002FVoiceCraft)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fjasonppy.github.io\u002FVoiceCraft_web\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLocalLLaMA\u002Fcomments\u002F1bmxfk3\u002Fvoicecraft_zeroshot_speech_editing_and\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FeikybOi8iwU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPJ2qSjycLcw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJxRrHpq-hys)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fjasonppy\u002FVoiceCraft\u002Fblob\u002Fmaster\u002Fvoicecraft-gradio-colab.ipynb) | 21.04.2024 |\n| ZeST | Method for zero-shot material transfer to an object in the input image given a material exemplar image | \u003Cul>\u003Cli>[Ta-Ying Cheng](https:\u002F\u002Fttchengab.github.io\u002F)\u003C\u002Fli> \u003Cli>[Prafull Sharma](https:\u002F\u002Fprafullsharma.net\u002F)\u003C\u002Fli> \u003Cli>[Andrew Markham](https:\u002F\u002Fwww.cs.ox.ac.uk\u002Fpeople\u002Fandrew.markham\u002F)\u003C\u002Fli> \u003Cli>[Niki Trigoni](https:\u002F\u002Fwww.cs.ox.ac.uk\u002Fpeople\u002Fniki.trigoni\u002F)\u003C\u002Fli> \u003Cli>[Varun Jampani](https:\u002F\u002Fvarunjampani.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fttchengab\u002Fzest_code?style=social)](https:\u002F\u002Fgithub.com\u002Fttchengab\u002Fzest_code) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.06425)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkealiu\u002FComfyUI-ZeroShot-MTrans)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fh94\u002FIP-Adapter), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fintel-isl\u002FDPT\u002Freleases\u002Fdownload\u002F1_0\u002Fdpt_hybrid-midas-501f0c75.pt)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fxthemadgenius.medium.com\u002Fzest-unlocks-material-magic-in-single-image-transfers-05f7ff7ee483)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fttchengab.github.io\u002Fzest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Flearnmachinelearning\u002Fcomments\u002F1c0wpjd\u002Fzest_zeroshot_material_transfer_from_a_single\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FatG1VvgeG_g)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002Fzest-jupyter\u002Fblob\u002Fmain\u002Fzest_jupyter.ipynb) | 16.04.2024 |\n| InstantMesh | Feed-forward framework for instant 3D mesh generation from a single image, featuring state-of-the-art generation quality and significant training scalability | \u003Cul>\u003Cli>[Jiale Xu](https:\u002F\u002Fgithub.com\u002Fbluestyle97)\u003C\u002Fli> \u003Cli>[Weihao Cheng](https:\u002F\u002Fwww.cheng.website\u002F)\u003C\u002Fli> \u003Cli>[Yiming Gao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=uRCc-McAAAAJ)\u003C\u002Fli> \u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Shenghua Gao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=fe-1v0MAAAAJ)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTencentARC\u002FInstantMesh?style=social)](https:\u002F\u002Fgithub.com\u002FTencentARC\u002FInstantMesh) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2404.07191)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdanielgatis\u002Frembg), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002F3DTopia\u002FOpenLRM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnv-tlabs\u002FFlexiCubes)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FTencentARC\u002FInstantMesh)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1c5hs3e\u002Finstantmesh_efficient_3d_mesh_generation_from_a\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBvngSJOStvQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FInstantMesh-jupyter\u002Fblob\u002Fmain\u002FInstantMesh_jupyter.ipynb) | 16.04.2024 |\n| Würstchen | Architecture for text-to-image synthesis that combines competitive performance with unprecedented cost-effectiveness for large-scale text-to-image diffusion models | \u003Cul>\u003Cli>[Pablo Pernias](https:\u002F\u002Fgithub.com\u002Fpabloppp)\u003C\u002Fli> \u003Cli>[Dominic Rampas](https:\u002F\u002Fgithub.com\u002Fdome272)\u003C\u002Fli> \u003Cli>[Mats Richter](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=xtlV5SAAAAAJ)\u003C\u002Fli> \u003Cli>[Christopher Pal](https:\u002F\u002Fwww.polymtl.ca\u002Fexpertises\u002Fpal-christopher-j)\u003C\u002Fli> \u003Cli>[Marc Aubreville](https:\u002F\u002Flme.tf.fau.de\u002Fperson\u002Faubreville\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdome272\u002Fwuerstchen?style=social)](https:\u002F\u002Fgithub.com\u002Fdome272\u002Fwuerstchen) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.00637)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fwuerstchen)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F16hsklt\u002Fw%C3%BCrstchen_is_here_a_game_changing_fastest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FogJsCPqgFMk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdome272\u002FWuerstchen\u002Fblob\u002Fmain\u002Fw%C3%BCrstchen-stage-C.ipynb) | 06.04.2024 |\n| BEiT | Self-supervised vision representation model, which stands for Bidirectional Encoder representation from Image Transformers | \u003Cul>\u003Cli>[Hangbo Bao](https:\u002F\u002Faddf400.github.io\u002F)\u003C\u002Fli> \u003Cli>[Li Dong](https:\u002F\u002Fdong.li\u002F)\u003C\u002Fli> \u003Cli>[Songhao Piao](https:\u002F\u002Fhomepage.hit.edu.cn\u002Fpiaosh)\u003C\u002Fli> \u003Cli>[Furu Wei](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Ffuwei\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fc7c1b3c2320.png)](https:\u002F\u002Fdoi.org\u002F10.48550\u002FarXiv.2106.08254) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002Funilm?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Funilm\u002Ftree\u002Fmaster\u002Fbeit) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.10442), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.06366), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.01127)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fapex), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdeit), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fpytorch-image-models), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdinov)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ftransformers\u002Fmaster\u002Fmodel_doc\u002Fbeit.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Freview-beit-bert-pre-training-of-image-transformers-c14a7ef7e295)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fsemantic-segmentation-on-ade20k)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVEEUklIWuNE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FSYEIvkPbyhU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FNielsRogge\u002FTransformers-Tutorials\u002Fblob\u002Fmaster\u002FBEiT\u002FUnderstanding_BeitForMaskedImageModeling.ipynb) | 30.03.2024 |\n| AudioSep | Foundation model for open-domain audio source separation with natural language queries | \u003Cul>\u003Cli>[Xubo Liu](https:\u002F\u002Fliuxubo717.github.io\u002F)\u003C\u002Fli> \u003Cli>[Qiuqiang Kong](https:\u002F\u002Fqiuqiangkong.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yan Zhao](https:\u002F\u002Fcliffzhao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haohe Liu](https:\u002F\u002Fhaoheliu.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yi Yuan](https:\u002F\u002Fwww.surrey.ac.uk\u002Fpeople\u002Fyi-yuan)\u003C\u002Fli> \u003Cli>[Yuzhuo Liu](https:\u002F\u002Fgithub.com\u002Fredrabbit94)\u003C\u002Fli> \u003Cli>[Rui Xia](https:\u002F\u002Fscholar.google.co.uk\u002Fcitations?user=26oErxwAAAAJ)\u003C\u002Fli> \u003Cli>[Yuxuan Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=3RaOfJkAAAAJ)\u003C\u002Fli> \u003Cli>[Mark Plumbley](https:\u002F\u002Fwww.surrey.ac.uk\u002Fpeople\u002Fmark-plumbley)\u003C\u002Fli> \u003Cli>[Wenwu Wang](http:\u002F\u002Fpersonal.ee.surrey.ac.uk\u002FPersonal\u002FW.Wang\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAudio-AGI\u002FAudioSep?style=social)](https:\u002F\u002Fgithub.com\u002FAudio-AGI\u002FAudioSep) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.05037)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Faudio-agi.github.io\u002FSeparate-Anything-You-Describe\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAudio-AGI\u002FAudioSep\u002Fblob\u002Fmain\u002FAudioSep_Colab.ipynb) | 15.03.2024 |\n| AQLM | Extreme Compression of Large Language Models via Additive Quantization | \u003Cul>\u003Cli>[Vage Egiazarian](https:\u002F\u002Fgithub.com\u002FVahe1994)\u003C\u002Fli> \u003Cli>[Andrei Panferov](https:\u002F\u002Fblog.panferov.org\u002F)\u003C\u002Fli> \u003Cli>[Denis Kuznedelev](https:\u002F\u002Fgithub.com\u002FGodofnothing)\u003C\u002Fli> \u003Cli>[Elias Frantar](https:\u002F\u002Fefrantar.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Artem Babenko](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2Kv3JP0AAAAJ)\u003C\u002Fli> \u003Cli>[Dan Alistarh](https:\u002F\u002Fgithub.com\u002Fdalistarh)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FVahe1994\u002FAQLM?style=social)](https:\u002F\u002Fgithub.com\u002FVahe1994\u002FAQLM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.06118)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fdatasets\u002Fmain\u002Fen\u002Fcache#cache-directory), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Ftogethercomputer\u002FRedPajama-Data-1T-Sample), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002FVahe1994\u002FAQLM)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLearningMachines\u002Fcomments\u002F1atvrnl\u002F240106118_extreme_compression_of_large_language\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQx8PNk4OkUA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhAHBKAXO-88)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FVahe1994\u002FAQLM\u002Fblob\u002Fmain\u002Fnotebooks\u002Fcolab_example.ipynb) | 08.03.2024 |\n| YOLOv9 | Learning What You Want to Learn Using Programmable Gradient Information | \u003Cul>\u003Cli>[Chien-Yao Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=DkQh4M4AAAAJ)\u003C\u002Fli> \u003Cli>[I-Hau Yeh](https:\u002F\u002Fieeexplore.ieee.org\u002Fauthor\u002F37088448531)\u003C\u002Fli> \u003Cli>[Hong-Yuan Mark Liao](https:\u002F\u002Fhomepage.iis.sinica.edu.tw\u002Fpages\u002Fliao\u002Findex_zh.html)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FWongKinYiu\u002Fyolov9?style=social)](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002Fyolov9) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.13616), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.16921)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Flearnopencv.com\u002Fyolov9-advancing-the-yolo-legacy\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002Fyolor), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FVDIGPKU\u002FDynamicDet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDingXiaoH\u002FRepVGG)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fkadirnar\u002FYolov9), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fmerve\u002Fyolov9)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@Mert.A\u002Fhow-to-use-yolov9-for-object-detection-93598ad88d7d)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FXHT2c8jT3Bc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3iLJ6YWPg28), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fdccf_sJF0Gg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Froboflow-ai\u002Fnotebooks\u002Fblob\u002Fmain\u002Fnotebooks\u002Ftrain-yolov9-object-detection-on-custom-dataset.ipynb) | 05.03.2024 |\n| Multi-LoRA Composition | LoRA Switch and LoRA Composite, approaches that aim to surpass traditional techniques in terms of accuracy and image quality, especially in complex compositions | \u003Cul>\u003Cli>[Ming Zhong](https:\u002F\u002Fmaszhongming.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yelong Shen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=S6OFEFEAAAAJ)\u003C\u002Fli> \u003Cli>[Shuohang Wang](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fshuowa\u002F)\u003C\u002Fli> \u003Cli>[Yadong Lu](https:\u002F\u002Fadamlu123.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yizhu Jiao](https:\u002F\u002Fyzjiao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Siru Ouyang](https:\u002F\u002Fozyyshr.github.io\u002F)\u003C\u002Fli> \u003Cli>[Donghan Yu](https:\u002F\u002Fplusross.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiawei Han](https:\u002F\u002Fhanj.cs.illinois.edu\u002F)\u003C\u002Fli> \u003Cli>[Weizhu Chen](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fwzchen\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmaszhongming\u002FMulti-LoRA-Composition?style=social)](https:\u002F\u002Fgithub.com\u002Fmaszhongming\u002FMulti-LoRA-Composition) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2402.16843)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@letscodeai\u002Fmulti-lora-composition-for-image-generation-f2706528c590)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fninjasaid13\u002Fcomments\u002F1b13q8s\u002Fmultilora_composition_for_image_generation\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fx.com\u002FMingZhong_\u002Fstatus\u002F1762347881812443575?s=20)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fmaszhongming.github.io\u002FMulti-LoRA-Composition\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1eSTj6qGOtSY5NaazwwN3meXOzEZxgaZq) | 03.03.2024 |\n| AMARETTO | Multiscale and multimodal inference of regulatory networks to identify cell circuits and their drivers shared and distinct within and across biological systems of human disease | \u003Cul>\u003Cli>[Nathalie Pochet](http:\u002F\u002Fportals.broadinstitute.org\u002Fpochetlab\u002F)\u003C\u002Fli> \u003Cli>[Olivier Gevaert](https:\u002F\u002Fprofiles.stanford.edu\u002Folivier-gevaert)\u003C\u002Fli> \u003Cli>[Mohsen Nabian](https:\u002F\u002Fgithub.com\u002Fmonabiyan)\u003C\u002Fli> \u003Cli>[Jayendra Shinde](https:\u002F\u002Fjayendrashinde91.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Celine Everaert](http:\u002F\u002Fwww.crig.ugent.be\u002Fen\u002Fnode\u002F510)\u003C\u002Fli> \u003Cli>[Thorin Tabor](http:\u002F\u002Fthorin.tabcreations.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgevaertlab\u002FAMARETTO?style=social)](https:\u002F\u002Fgithub.com\u002Fgevaertlab\u002FAMARETTO) \u003Cul>\u003Cli>[bioconductor](https:\u002F\u002Fbioconductor.org\u002Fpackages\u002Frelease\u002Fbioc\u002Fhtml\u002FAMARETTO.html)\u003C\u002Fli>\u003Cli>[project](http:\u002F\u002Fportals.broadinstitute.org\u002Fpochetlab\u002Famaretto.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1JfnRoNgTVX_7VEGAAmjGjwP_yX2tdDxs) | 28.02.2024 |\n| LIDA | Tool for generating grammar-agnostic visualizations and infographics | [Victor Dibia](https:\u002F\u002Fvictordibia.com\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_14e441d2aa9f.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002F2023.acl-demo.11) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002Flida?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Flida) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.02927)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvictordibia\u002Fllmx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flida-project\u002Flida-streamlit)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@c17hawke\u002Flida-automatically-generate-visualization-and-with-llms-the-future-of-data-visualization-6bc556876b46)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmicrosoft.github.io\u002Flida\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FexYi9W-dhME), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FU9K1Cu45nMQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F6xcCwlDx6f8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmicrosoft\u002Flida\u002Fblob\u002Fmain\u002Fnotebooks\u002Ftutorial.ipynb) | 06.02.2024 |\n| ViT | Vision Transformer and MLP-Mixer Architectures | \u003Cul>\u003Cli>[Alexey Dosovitskiy](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=FXNJRDoAAAAJ)\u003C\u002Fli> \u003Cli>[Lucas Beyer](http:\u002F\u002Flucasb.eyer.be)\u003C\u002Fli> \u003Cli>[Alexander Kolesnikov](https:\u002F\u002Fgithub.com\u002Fakolesnikoff)\u003C\u002Fli> \u003Cli>[Dirk Weissenborn](https:\u002F\u002Fgithub.com\u002Fdirkweissenborn)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xiaohua Zhai](https:\u002F\u002Fgithub.com\u002Fxiaohuazhai)\u003C\u002Fli> \u003Cli>[Thomas Unterthiner](https:\u002F\u002Fgithub.com\u002Funtom)\u003C\u002Fli> \u003Cli>[Mostafa Dehghani](https:\u002F\u002Fwww.mostafadehghani.com\u002F)\u003C\u002Fli> \u003Cli>[Matthias Minderer](https:\u002F\u002Fmatthias.minderer.net\u002F)\u003C\u002Fli> \u003Cli>[Georg Heigold](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=WwqlChAAAAAJ)\u003C\u002Fli> \u003Cli>[Sylvain Gelly](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=m7LvuTkAAAAJ)\u003C\u002Fli> \u003Cli>[Jakob Uszkoreit](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=mOG0bwsAAAAJ)\u003C\u002Fli> \u003Cli>[Neil Houlsby](https:\u002F\u002Fneilhoulsby.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fvision_transformer?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fvision_transformer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.11929), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.01601), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.01601), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.10270), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.01548), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.07991), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.08065)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fblog.research.google\u002F2022\u002F04\u002Flocked-image-tuning-adding-language.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fpytorch-image-models), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fflaxformer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fmodels)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@weiwen21\u002Fan-image-is-worth-16x16-words-transformers-for-image-recognition-at-scale-957f88e53726)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTrdevFK_am4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FHZ4j_U3FC94), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F7K4Z8RqjWIk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FoDtcobGQ7xU?si=C2EgZTESzhTXFSq6), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fv6xj_DG-UEo)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Fvision_transformer\u002Fblob\u002Fmain\u002Fvit_jax.ipynb) | 06.02.2024 |\n| Qwen | Comprehensive language model series that encompasses distinct models with varying parameter counts | [qwenlm](https:\u002F\u002Fqwenlm.github.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FQwenLM\u002FQwen?style=social)](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.16609), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.09685), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.14314), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2307.11088)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FCV4E9rpNSD)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fqwenllm\u002Fqwen)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen2.5), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen-Agent), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FQwenLM\u002Fqwen.cpp), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDao-AILab\u002Fflash-attention), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAutoGPTQ\u002FAutoGPTQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FDeepSpeed)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FQwen)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fpytorch.org\u002Fdocs\u002Fstable\u002Felastic\u002Frun.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fy6Wh4SpRoao), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbJmx_fAOW78), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FALArhCnz8rY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FQwenLM\u002FQwen\u002Fblob\u002Fmaster\u002Frecipes\u002Fquickstart\u002Fqwen.ipynb) | 30.01.2024 |\n| 3D Ken Burns | A reference implementation of 3D Ken Burns Effect from a Single Image using PyTorch - given a single input image, it animates this still image with a virtual camera scan and zoom subject to motion parallax | [Manuel Romero](https:\u002F\u002Fmrm8488.github.io\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ff89b6fc991e.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3355089.3356528) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsniklaus\u002F3d-ken-burns?style=social)](https:\u002F\u002Fgithub.com\u002Fsniklaus\u002F3d-ken-burns) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.05483)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=WrajxHHfRBA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmrm8488\u002Fshared_colab_notebooks\u002Fblob\u002Fmaster\u002F3D_Ken_Burns.ipynb) | 24.01.2024 |\n| VALL-E X | Cross-lingual neural codec language model for cross-lingual speech synthesis | \u003Cul>\u003Cli>[Ziqiang Zhang](https:\u002F\u002Fgithub.com\u002Fonisac-K)\u003C\u002Fli> \u003Cli>[Long Zhou](https:\u002F\u002Flong-zhou.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chengyi Wang](https:\u002F\u002Fcywang97.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sanyuan Chen](https:\u002F\u002Fsanyuan-chen.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yu Wu](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fyuwu1\u002F)\u003C\u002Fli> \u003Cli>[Shujie Liu](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fshujliu\u002F)\u003C\u002Fli> \u003Cli>[Zhuo Chen](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fzhuc\u002F)\u003C\u002Fli> \u003Cli>[Yanqing Liu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=dIJFz4UAAAAJ)\u003C\u002Fli> \u003Cli>[Huaming Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=aJDLg5IAAAAJ)\u003C\u002Fli> \u003Cli>[Jinyu Li](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fjinyli\u002F)\u003C\u002Fli> \u003Cli>[Lei He](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=EKl9yY8AAAAJ)\u003C\u002Fli> \u003Cli>[Sheng Zhao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=689bIIwAAAAJ)\u003C\u002Fli> \u003Cli>[Furu Wei](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Ffuwei\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPlachtaa\u002FVALL-E-X?style=social)](https:\u002F\u002Fgithub.com\u002FPlachtaa\u002FVALL-E-X) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.03926), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.02111), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.03143)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fplachtaa.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FqCBRmAnTxg)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flifeiteng\u002Fvall-e)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FPlachta\u002FVALL-E-X)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fsyncedreview\u002Fspeak-a-foreign-language-in-your-own-voice-1dafa42f78d9)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fproject\u002Fvall-e-x)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F7qgfoVFQmvk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1yyD_sz531QntLKowMHo-XxorsFBCfKul) | 19.01.2024 |\n| PhotoMaker | Efficient personalized text-to-image generation method, which mainly encodes an arbitrary number of input ID images into a stack ID embedding for preserving ID information | \u003Cul>\u003Cli>[Zhen Li](https:\u002F\u002Fpaper99.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mingdeng Cao](https:\u002F\u002Fgithub.com\u002Fljzycmd)\u003C\u002Fli> \u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhongang Qi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=zJvrrusAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ming-Ming Cheng](https:\u002F\u002Fmmcheng.net\u002Fcmm\u002F)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTencentARC\u002FPhotoMaker?style=social)](https:\u002F\u002Fgithub.com\u002FTencentARC\u002FPhotoMaker) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.04461)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbmaltais\u002FPhotoMaker), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsdbds\u002FPhotoMaker-for-windows), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZHO-ZHO-ZHO\u002FComfyUI-PhotoMaker), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Ffastcomposer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTencentARC\u002FT2I-Adapter), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftencent-ailab\u002FIP-Adapter)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FTencentARC\u002FPhotoMaker)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@christopheverdier\u002Fphotomaker-the-art-of-ai-consistent-characters-generation-cf2cd037bc3e)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fphoto-maker.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F197bfj9\u002Ftencentarc_releases_photomaker\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FNWIdzTEk5O4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FZTck128jfFY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FTencentARC\u002FPhotoMaker\u002Fblob\u002Fmain\u002Fphotomaker_demo.ipynb) | 18.01.2024 |\n| DDColor | End-to-end method with dual decoders for image colorization | \u003Cul>\u003Cli>[Xiaoyang Kang](https:\u002F\u002Fpiddnad.github.io\u002Fxiaoyangkang)\u003C\u002Fli> \u003Cli>[Tao Yang](https:\u002F\u002Fcg.cs.tsinghua.edu.cn\u002Fpeople\u002F~tyang\u002F)\u003C\u002Fli> \u003Cli>[Wenqi Ouyang](https:\u002F\u002Fvicky0522.github.io\u002FWenqi-Ouyang\u002F)\u003C\u002Fli> \u003Cli>[Peiran Ren](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=x5dEuxsAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lingzhi Li](https:\u002F\u002Flingzhili.com\u002F)\u003C\u002Fli> \u003Cli>[Xuansong Xie](https:\u002F\u002Fgithub.com\u002Fxungie)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fpiddnad\u002FDDColor?style=social)](https:\u002F\u002Fgithub.com\u002Fpiddnad\u002FDDColor) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.11613)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjixiaozhong\u002FColorFormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FKIMGEONUNG\u002FBigColor)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FDDColor-colab\u002Fblob\u002Fmain\u002FDDColor_colab.ipynb) | 15.01.2024 |\n| PASD | Pixel-aware stable diffusion network to achieve robust Real-ISR as well as personalized stylization | \u003Cul>\u003Cli>[Tao Yang](https:\u002F\u002Fcg.cs.tsinghua.edu.cn\u002Fpeople\u002F~tyang)\u003C\u002Fli> \u003Cli>[Peiran Ren](http:\u002F\u002Frenpr.org\u002F)\u003C\u002Fli> \u003Cli>[Xuansong Xie](https:\u002F\u002Fgithub.com\u002Fxungie)\u003C\u002Fli> \u003Cli>[Lei Zhang](https:\u002F\u002Fwww4.comp.polyu.edu.hk\u002F~cslzhang)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyangxy\u002FPASD?style=social)](https:\u002F\u002Fgithub.com\u002Fyangxy\u002FPASD) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.14469)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fpkuliyi2015\u002Fmultidiffusion-upscaler-for-automatic1111)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Frunwayml\u002Fstable-diffusion-v1-5), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fnitrosocke\u002Fmo-di-diffusion)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F18qxe5q\u002Fpixelaware_stable_diffusion_for_realistic_image\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1lZ_-rSGcmreLCiRniVT973x6JLjFiC-b) | 12.01.2024 |\n| HandRefiner | Refining Malformed Hands in Generated Images by Diffusion-based Conditional Inpainting | \u003Cul>\u003Cli>[Wenquan Lu](https:\u002F\u002Fgithub.com\u002Fwenquanlu)\u003C\u002Fli> \u003Cli>[Yufei Xu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=hlYWxX8AAAAJ)\u003C\u002Fli> \u003Cli>[Jing Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=9jH5v74AAAAJ)\u003C\u002Fli> \u003Cli>[Chaoyue Wang](https:\u002F\u002Fwang-chaoyue.github.io\u002F)\u003C\u002Fli> \u003Cli>[Dacheng Tao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RwlJNLcAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwenquanlu\u002FHandRefiner?style=social)](https:\u002F\u002Fgithub.com\u002Fwenquanlu\u002FHandRefiner) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.17957)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFannovel16\u002Fcomfyui_controlnet_aux), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMikubill\u002Fsd-webui-controlnet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FMeshGraphormer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1881z4v\u002Fhandrefiner_refining_malformed_hands_in_generated\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTt-Fyn1RA6c)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FHandRefiner-colab\u002Fblob\u002Fmain\u002FHandRefiner_colab.ipynb) | 08.01.2024 |\n| LLaVA | Large Language and Vision Assistant, an end-to-end trained large multimodal model that connects a vision encoder and LLM for general-purpose visual and language understanding | \u003Cul>\u003Cli>[Haotian Liu](https:\u002F\u002Fhliu.cc\u002F)\u003C\u002Fli> \u003Cli>[Chunyuan Li](https:\u002F\u002Fchunyuan.li\u002F)\u003C\u002Fli> \u003Cli>[Qingyang Wu](https:\u002F\u002Fqywu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yong Jae Lee](https:\u002F\u002Fpages.cs.wisc.edu\u002F~yongjaelee\u002F)\u003C\u002Fli> \u003Cli>[Yuheng Li](https:\u002F\u002Fyuheng-li.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhaotian-liu\u002FLLaVA?style=social)](https:\u002F\u002Fgithub.com\u002Fhaotian-liu\u002FLLaVA) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.08485), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.03744), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.00890), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.09958), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.14895)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fllava.hliu.cc\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fggerganov\u002Fllama.cpp\u002Fpull\u002F3436), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FLLaVA-Med), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FUX-Decoder\u002FSegment-Everything-Everywhere-All-At-Once), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FLuodian\u002FOtter), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FInstruction-Tuning-with-GPT-4\u002FGPT-4-LLM)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fliuhaotian\u002FLLaVA-Pretrain), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fliuhaotian\u002FLLaVA-Pretrained-Projectors)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fxthemadgenius.medium.com\u002Fhow-to-use-llava-large-language-and-vision-assistant-732c666b5ed0)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fllava-vl.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FmkI7EPD1vp8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fkx1VpI6JzsY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRxBSmbdJ1I8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FmdYycY4lsuE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Ft7I46dxfmWs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FKRAQkJC-XJU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FLLaVA-colab\u002Fblob\u002Fmain\u002FLLaVA_13b_4bit_vanilla_colab.ipynb) | 22.12.2023 |\n| Background Matting V2 | Real-time, high-resolution background replacement technique which operates at 30fps in 4K resolution, and 60fps for HD on a modern GPU | \u003Cul>\u003Cli>[Shanchuan Lin](https:\u002F\u002Fgithub.com\u002FPeterL1n)\u003C\u002Fli> \u003Cli>[Andrey Ryabtsev](https:\u002F\u002Fgithub.com\u002Fandreyryabtsev)\u003C\u002Fli> \u003Cli>[Soumyadip Sengupta](https:\u002F\u002Fgithub.com\u002Fsenguptaumd)\u003C\u002Fli> \u003Cli>[Brian Curless](https:\u002F\u002Fhomes.cs.washington.edu\u002F~curless\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Steve Seitz](https:\u002F\u002Fwww.smseitz.com\u002F)\u003C\u002Fli> \u003Cli>[Ira Kemelmacher-Shlizerman](https:\u002F\u002Fwww.irakemelmacher.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_27ef0e6ed9d2.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00865) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPeterL1n\u002FBackgroundMattingV2?style=social)](https:\u002F\u002Fgithub.com\u002FPeterL1n\u002FBackgroundMattingV2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.07810)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsenguptaumd\u002FBackground-Matting), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fandreyryabtsev\u002FBGMv2-webcam-plugin-linux)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgrail.cs.washington.edu\u002Fprojects\u002Fbackground-matting-v2\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FoMfPTeYDF9g), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fb7ps21MVyTA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1cTxFq1YuoJ5QPqaTcnskwlHDolnjBkB9) | 22.12.2023 |\n| FreeInit | Concise yet effective method to improve temporal consistency of videos generated by diffusion modelsconcise yet effective method to improve temporal consistency of videos generated by diffusion models | \u003Cul>\u003Cli>[Tianxing Wu](https:\u002F\u002Ftianxingwu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chenyang Si](http:\u002F\u002Fchenyangsi.top\u002F)\u003C\u002Fli> \u003Cli>[Yuming Jiang](https:\u002F\u002Fyumingj.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ziqi Huang](https:\u002F\u002Fziqihuangg.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cba6c3f73678.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-72646-0_22) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTianxingWu\u002FFreeInit?style=social)](https:\u002F\u002Fgithub.com\u002FTianxingWu\u002FFreeInit) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.07537)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FKosinkadink\u002FComfyUI-AnimateDiff-Evolved)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FTianxingWu\u002FFreeInit)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftianxingwu.github.io\u002Fpages\u002FFreeInit\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FlS5IYbAqriI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FFreeInit-colab\u002Fblob\u002Fmain\u002FFreeInit_colab.ipynb) | 21.12.2023 |\n| Gaussian Splatting | State-of-the-art visual quality while maintaining competitive training times and importantly allow high-quality real-time (≥ 100 fps) novel-view synthesis at 1080p resolution | \u003Cul>\u003Cli>[Bernhard Kerbl](https:\u002F\u002Fwww.cg.tuwien.ac.at\u002Fstaff\u002FBernhardKerbl)\u003C\u002Fli> \u003Cli>[Georgios Kopanas](https:\u002F\u002Fgrgkopanas.github.io\u002F)\u003C\u002Fli> \u003Cli>[Thomas Leimkühler](https:\u002F\u002Fpeople.mpi-inf.mpg.de\u002F~tleimkue\u002F)\u003C\u002Fli> \u003Cli>[George Drettakis](http:\u002F\u002Fwww-sop.inria.fr\u002Fmembers\u002FGeorge.Drettakis\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0c34c9f6bf85.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3592433) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgraphdeco-inria\u002Fgaussian-splatting?style=social)](https:\u002F\u002Fgithub.com\u002Fgraphdeco-inria\u002Fgaussian-splatting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.04079)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcamenduru\u002Fgaussian-splatting)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Faxinc-ai\u002F3d-gaussian-splatting-real-time-rendering-of-photorealistic-scenes-f7f1a47f060)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Frepo-sam.inria.fr\u002Ffungraph\u002F3d-gaussian-splatting\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fsingularity\u002Fcomments\u002F163jeqa\u002F3d_gaussian_splatting_for_realtime_radiance_field\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FT_kXY43VZnk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUXtuigy_wYc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FHVv_IQKlafQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fw43KV79LsFw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTLK3TDDcJFU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FkShNYOuDnlI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjuRMRej2d5c)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002Fgaussian-splatting-colab\u002Fblob\u002Fmain\u002Fgaussian_splatting_colab.ipynb) | 19.12.2023 |\n| SMPLer-X | Scaling up EHPS towards the first generalist foundation model, with up to ViT-Huge as the backbone and training with up to 4.5M instances from diverse data sources | \u003Cul>\u003Cli>[Zhongang Cai](https:\u002F\u002Fcaizhongang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Wanqi Yin](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=zlIJwBEAAAAJ)\u003C\u002Fli> \u003Cli>[Ailing Zeng](https:\u002F\u002Failingzeng.site\u002F)\u003C\u002Fli> \u003Cli>[Chen Wei](https:\u002F\u002Fgithub.com\u002FWei-Chen-hub)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Qingping Sun](https:\u002F\u002Fgithub.com\u002Fttxskk)\u003C\u002Fli> \u003Cli>[Yanjun Wang](https:\u002F\u002Fgithub.com\u002FWYJSJTU)\u003C\u002Fli> \u003Cli>[Hui En Pang](https:\u002F\u002Fpangyyyyy.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haiyi Mei](https:\u002F\u002Fhaiyi-mei.com\u002F)\u003C\u002Fli> \u003Cli>[Mingyuan Zhang](https:\u002F\u002Fmingyuan-zhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Lei Zhang](https:\u002F\u002Fwww.leizhang.org\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli> \u003Cli>[Lei Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jZH2IPYAAAAJ)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcaizhongang\u002FSMPLer-X?style=social)](https:\u002F\u002Fgithub.com\u002Fcaizhongang\u002FSMPLer-X) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.17448)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopen-mmlab\u002Fmmhuman3d\u002Fblob\u002Fmain\u002Fdocs\u002Fhuman_data.md), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmks0601\u002FHand4Whole_RELEASE), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FIDEA-Research\u002FOSX)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fneurips.cc\u002Fvirtual\u002F2023\u002Fposter\u002F73473)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fcaizhongang.com\u002Fprojects\u002FSMPLer-X\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fmachinelearningnews\u002Fcomments\u002F176c5z7\u002Fthis_ai_research_proposes_smplerx_a_generalist\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDepTqbPpVzY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FaFTGFInUnM4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FSMPLer-X-colab\u002Fblob\u002Fmain\u002FSMPLer_X_colab.ipynb) | 18.12.2023 |\n| DeepCache | Training-free paradigm that accelerates diffusion models from the perspective of model architecture | \u003Cul>\u003Cli>[Xinyin Ma](https:\u002F\u002Fhorseee.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gongfan Fang](https:\u002F\u002Ffangggf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xinchao Wang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fsitexinchaowang\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhorseee\u002FDeepCache?style=social)](https:\u002F\u002Fgithub.com\u002Fhorseee\u002FDeepCache) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.00858)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fdiffusers\u002Fv0.24.0\u002Fen\u002Fapi\u002Fpipelines\u002Fstable_diffusion\u002Ftext2img#diffusers.StableDiffusionPipeline)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fhorseee.github.io\u002FDiffusion_DeepCache\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F18b40hh\u002Fdeepcache_accelerating_diffusion_models_for_free\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FDeepCache-colab\u002Fblob\u002Fmain\u002FDeepCache_colab.ipynb) | 18.12.2023 |\n| MagicAnimate | Diffusion-based framework that aims at enhancing temporal consistency, preserving reference image faithfully, and improving animation fidelity | \u003Cul>\u003Cli>[Zhongcong Xu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=-4iADzMAAAAJ)\u003C\u002Fli> \u003Cli>[Jianfeng Zhang](http:\u002F\u002Fjeff95.me\u002F)\u003C\u002Fli> \u003Cli>[Jun Hao Liew](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=8gm-CYYAAAAJ)\u003C\u002Fli> \u003Cli>[Hanshu Yan](https:\u002F\u002Fhanshuyan.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jiawei Liu](https:\u002F\u002Fjia-wei-liu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chenxu Zhang](https:\u002F\u002Fzhangchenxu528.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiashi Feng](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fjshfeng\u002Fhome)\u003C\u002Fli> \u003Cli>[Mike Shou](https:\u002F\u002Fsites.google.com\u002Fview\u002Fshowlab)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagic-research\u002Fmagic-animate?style=social)](https:\u002F\u002Fgithub.com\u002Fmagic-research\u002Fmagic-animate) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.16498)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fzcxu-eric\u002FMagicAnimate), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Frunwayml\u002Fstable-diffusion-v1-5), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fstabilityai\u002Fsd-vae-ft-mse)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@AIWorldBlog\u002Frevolutionizing-image-animation-with-magicanimate-technology-78cc94151915)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fshowlab.github.io\u002Fmagicanimate\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fwww.magicanimate.org\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Ftd27SyA9M80), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F1pATjLFvNtY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FHeXknItbMM8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FMagicAnimate-colab\u002Fblob\u002Fmain\u002FMagicAnimate_colab.ipynb) | 18.12.2023 |\n| DiffBIR | Towards Blind Image Restoration with Generative Diffusion Prior | \u003Cul>\u003Cli>[Xinqi Lin](https:\u002F\u002Fgithub.com\u002F0x3f3f3f3fun)\u003C\u002Fli> \u003Cli>[Jingwen He](https:\u002F\u002Fgithub.com\u002Fhejingwenhejingwen)\u003C\u002Fli> \u003Cli>[Ziyan Chen](https:\u002F\u002Fgithub.com\u002Fziyannchen)\u003C\u002Fli> \u003Cli>[Zhaoyang Lyu](https:\u002F\u002Fzhaoyanglyu.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ben Fei](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=skQROj8AAAAJ)\u003C\u002Fli> \u003Cli>[Bo Dai](http:\u002F\u002Fdaibo.info\u002F)\u003C\u002Fli> \u003Cli>[Wanli Ouyang](https:\u002F\u002Fwlouyang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yu Qiao](https:\u002F\u002Fmmlab.siat.ac.cn\u002Fyuqiao)\u003C\u002Fli> \u003Cli>[Chao Dong](http:\u002F\u002Fxpixel.group\u002F2010\u002F01\u002F20\u002Fchaodong.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FXPixelGroup\u002FDiffBIR?style=social)](https:\u002F\u002Fgithub.com\u002FXPixelGroup\u002FDiffBIR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.15070)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Falbarji\u002Fmixture-of-diffusers)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fstabilityai\u002Fstable-diffusion-2-1-base)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002F0x3f3f3f3fun.github.io\u002Fprojects\u002Fdiffbir\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FrGnrpxWjBOg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FMIRiJGuGqsg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FDiffBIR-colab\u002Fblob\u002Fmain\u002FDiffBIR_colab.ipynb) | 18.12.2023 |\n| Segment and Track Anything | Framewoork that allows users to precisely and effectively segment and track any object in a video | \u003Cul>\u003Cli>[Yangming Cheng](https:\u002F\u002Fgithub.com\u002Fyamy-cheng)\u003C\u002Fli> \u003Cli>[Liulei Li](https:\u002F\u002Fgithub.com\u002FlingorX)\u003C\u002Fli> \u003Cli>[Yuanyou Xu](https:\u002F\u002Fgithub.com\u002Fyoxu515)\u003C\u002Fli> \u003Cli>[Xiaodi Li](https:\u002F\u002Fgithub.com\u002FLiNO3Dy)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Zongxin Yang](https:\u002F\u002Fz-x-yang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Wenguan Wang](https:\u002F\u002Fsites.google.com\u002Fview\u002Fwenguanwang)\u003C\u002Fli> \u003Cli>[Yi Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RMSuNFwAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fz-x-yang\u002FSegment-and-Track-Anything?style=social)](https:\u002F\u002Fgithub.com\u002Fz-x-yang\u002FSegment-and-Track-Anything) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.06558), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.02643), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002Feb890c36af87e4ca82e8ef7bcba6a284-Abstract-Conference.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyoxu515\u002Faot-benchmark)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FShilongLiu\u002FGroundingDINO\u002Fresolve\u002Fmain\u002Fgroundingdino_swint_ogc.pth)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002Feb890c36af87e4ca82e8ef7bcba6a284-Abstract-Conference.html), [\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2021\u002Fhash\u002F147702db07145348245dc5a2f2fe5683-Abstract.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDF0iFSsX8KY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUJvKPng9_DA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fm1oFavjIaCM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUPhtpf1k6HA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FXyd54AngvV8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FeZrdna8JkoQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhPjw28Ul4cw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fl7hXM1a3nEA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FnXfq17X6ohk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5oieHqFIJPc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FcK5MPFdJdSY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUFtwFaOfx2I)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1R10N70AJaslzADFqb-a5OihYkllWEVxB) | 08.12.2023 |\n| AudioLDM | Text-to-audio system that is built on a latent space to learn the continuous audio representations from contrastive language-audio pretraining latents | \u003Cul>\u003Cli>[Haohe Liu](https:\u002F\u002Fhaoheliu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zehua Chen](https:\u002F\u002Fgithub.com\u002FzehuachenImperial)\u003C\u002Fli> \u003Cli>[Yi Yuan](https:\u002F\u002Fwww.surrey.ac.uk\u002Fpeople\u002Fyi-yuan)\u003C\u002Fli> \u003Cli>[Xinhao Mei](https:\u002F\u002Fxinhaomei.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xubo Liu](https:\u002F\u002Fliuxubo717.github.io\u002F)\u003C\u002Fli> \u003Cli>[Danilo Mandic](https:\u002F\u002Fwww.imperial.ac.uk\u002Fpeople\u002Fd.mandic)\u003C\u002Fli> \u003Cli>[Wenwu Wang](http:\u002F\u002Fpersonal.ee.surrey.ac.uk\u002FPersonal\u002FW.Wang\u002F)\u003C\u002Fli> \u003Cli>[Mark Plumbley](https:\u002F\u002Fwww.surrey.ac.uk\u002Fpeople\u002Fmark-plumbley)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhaoheliu\u002FAudioLDM?style=social)](https:\u002F\u002Fgithub.com\u002Fhaoheliu\u002FAudioLDM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.12503)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FLAION-AI\u002FCLAP), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCompVis\u002Fstable-diffusion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftoshas\u002Ftorch-fidelity)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Faudioldm.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_0VTltNYhao)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Folaviinha\u002FNeuralTextToAudio\u002Fblob\u002Fmain\u002FAudioLDM_pub.ipynb) | 02.12.2023 |\n| TabPFN | Neural network that learned to do tabular data prediction | \u003Cul>\u003Cli>[Noah Hollmann](https:\u002F\u002Fgithub.com\u002Fnoahho)\u003C\u002Fli> \u003Cli>[Samuel Müller](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=pevYEjAAAAAJ)\u003C\u002Fli> \u003Cli>[Katharina Eggensperger](https:\u002F\u002Fgithub.com\u002FKEggensperger)\u003C\u002Fli> \u003Cli>[Frank Hutter](https:\u002F\u002Fml.informatik.uni-freiburg.de\u002Fprofile\u002Fhutter\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fautoml\u002FTabPFN?style=social)](https:\u002F\u002Fgithub.com\u002Fautoml\u002FTabPFN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.01848), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.11189), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.01342), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.03253), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.11189), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10510)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fwww.automl.org\u002Ftabpfn-a-transformer-that-solves-small-tabular-classification-problems-in-a-second\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002Ftunguz\u002Fstatus\u002F1578730907711655937)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBGTO5N5-ack)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F194mCs6SEPEW6C0rcP7xWzcEtt1RBc8jJ) | 29.11.2023 |\n| Concept Sliders | Plug-and-play low rank adaptors applied on top of pretrained models | \u003Cul>\u003Cli>[Rohit Gandikota](https:\u002F\u002Frohitgandikota.github.io\u002F)\u003C\u002Fli> \u003Cli>[Joanna Materzyńska](https:\u002F\u002Fjoaanna.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tingrui Zhou](https:\u002F\u002Fwww.p1at.dev\u002F)\u003C\u002Fli> \u003Cli>[Antonio Torralba](https:\u002F\u002Fgroups.csail.mit.edu\u002Fvision\u002Ftorralbalab\u002F)\u003C\u002Fli> \u003Cli>[David Bau](https:\u002F\u002Fbaulab.info\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Frohitgandikota\u002Fsliders?style=social)](https:\u002F\u002Fgithub.com\u002Frohitgandikota\u002Fsliders) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2311.12092), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.12598)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@furkangozukara\u002Fconcept-sliders-lora-adaptors-for-precise-control-in-diffusion-models-b7f6b36fabee)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2020\u002Fhash\u002F49856ed476ad01fcff881d57e161d73f-Abstract.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsliders.baulab.info\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F180zon7\u002Fconcept_sliders_lora_adaptors_for_precise_control\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Frohitgandikota\u002Fsliders\u002Fblob\u002Fmain\u002Fdemo_concept_sliders.ipynb) | 26.11.2023 |\n| Qwen-VL | Set of large-scale vision-language models designed to perceive and understand both text and images | \u003Cul>\u003Cli>[Jinze Bai](https:\u002F\u002Fgithub.com\u002Fjinze1994)\u003C\u002Fli> \u003Cli>[Shuai Bai](https:\u002F\u002Fgithub.com\u002FShuaiBai623)\u003C\u002Fli> \u003Cli>[Shusheng Yang](https:\u002F\u002Fshushengyang.com\u002F)\u003C\u002Fli> \u003Cli>[Shijie Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=DuAqyTwAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Sinan Tan](https:\u002F\u002Fwww.tinytangent.com\u002F)\u003C\u002Fli> \u003Cli>[Peng Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=7fjqA0YAAAAJ)\u003C\u002Fli> \u003Cli>[Junyang Lin](https:\u002F\u002Fjustinlin610.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chang Zhou](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QeSoG3sAAAAJ)\u003C\u002Fli> \u003Cli>[Jingren Zhou](http:\u002F\u002Fwww.cs.columbia.edu\u002F~jrzhou\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FQwenLM\u002FQwen-VL?style=social)](https:\u002F\u002Fgithub.com\u002FQwenLM\u002FQwen-VL) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2308.12966), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.09685), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.14314)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fmodelscope.cn\u002Fstudios\u002Fqwen\u002FQwen-VL-Chat-Demo\u002Fsummary)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002Fz3GAxXZ9Ce)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FBradyFU\u002FAwesome-Multimodal-Large-Language-Models\u002Ftree\u002FEvaluation), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOFA-Sys\u002FTouchStone), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FPanQiWei\u002FAutoGPTQ)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FAILab-CVC\u002FSEED-Bench_Leaderboard), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FQwen\u002FQwen-VL)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FElrSJDg23Po), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FE3MS8GfGWj4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fju09YaO7BGA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FQwen-VL-Chat-colab\u002Fblob\u002Fmain\u002FQwen_VL_Chat_colab.ipynb) | 24.11.2023 |\n| AnimeGANv3 | Double-tail generative adversarial network for fast photo animation | \u003Cul>\u003Cli>[Gang Liu](https:\u002F\u002Fgithub.com\u002Flg0061408)\u003C\u002Fli> \u003Cli>[Xin Chen](https:\u002F\u002Fgithub.com\u002FTachibanaYoshino)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c2bb7f34e2f5.png)](http:\u002F\u002Fdoi.org\u002F10.1587\u002Ftransinf.2023EDP7061) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTachibanaYoshino\u002FAnimeGANv3?style=social)](https:\u002F\u002Fgithub.com\u002FTachibanaYoshino\u002FAnimeGANv3) \u003Cul>\u003Cli>[project](https:\u002F\u002Ftachibanayoshino.github.io\u002FAnimeGANv3\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FEosubeJmAnE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5qLUflWb45E), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FiFjiaPlhVm4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FvJqQQMRYKh0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0KaScDxgyBw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F6WXhjXb5a-o)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1XYNWwM8Xq-U7KaTOqNap6A-Yq1f-V-FB) | 23.11.2023 |\n| Ithaca | First Deep Neural Network for the textual restoration, geographical and chronological attribution of ancient Greek inscriptions | \u003Cul>\u003Cli>[Yannis Assael](https:\u002F\u002Fwww.assael.gr\u002F)\u003C\u002Fli> \u003Cli>[Thea Sommerschield](https:\u002F\u002Ftheasommerschield.it\u002F)\u003C\u002Fli> \u003Cli>[Brendan Shillingford](https:\u002F\u002Fgithub.com\u002Fbshillingford)\u003C\u002Fli> \u003Cli>[Mahyar Bordbar](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=KB3ldSQAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[John Pavlopoulos](https:\u002F\u002Fipavlopoulos.github.io\u002F)\u003C\u002Fli> \u003Cli>[Marita Chatzipanagiotou](https:\u002F\u002Fgr.linkedin.com\u002Fin\u002Fmarita-chatzipanagiotou-b0611a1a2)\u003C\u002Fli> \u003Cli>[Ion Androutsopoulos](https:\u002F\u002Fpages.aueb.gr\u002Fusers\u002Fion\u002F)\u003C\u002Fli> \u003Cli>[Jonathan Prag](https:\u002F\u002Fwww.classics.ox.ac.uk\u002Fpeople\u002Fdr-jonathan-prag)\u003C\u002Fli> \u003Cli>[Nando de Freitas](https:\u002F\u002Fwww.cs.ox.ac.uk\u002Fpeople\u002Fnando.defreitas\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cac39cffe76a.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41586-022-04448-z) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Fithaca?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fithaca) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.06262)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsommerschield\u002Fiphi)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fodsc.medium.com\u002Fdeep-neural-networks-could-be-key-to-ancient-text-restoration-and-attribution-research-shows-81a2d89d9413), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fsyncedreview\u002Fithaca-paper-published-in-nature-the-first-dnn-designed-for-textual-restoration-and-geographical-ef395d56697e)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fithaca.deepmind.com\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Ftgeo0q\u002Fr_restoring_and_attributing_ancient_texts_using\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Fithaca\u002Fblob\u002Fmaster\u002Fcolabs\u002Fithaca_inference.ipynb) | 21.11.2023 |\n| PixArt-Σ | Weak-to-Strong Training of Diffusion Transformer for 4K Text-to-Image Generation | \u003Cul>\u003Cli>[Junsong Chen](https:\u002F\u002Flawrence-cj.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chongjian Ge](https:\u002F\u002Fchongjiange.github.io\u002F)\u003C\u002Fli> \u003Cli>[Enze Xie](https:\u002F\u002Fxieenze.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yue Wu](https:\u002F\u002Fyuewuhkust.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lewei Yao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=hqDyTg8AAAAJ)\u003C\u002Fli> \u003Cli>[Xiaozhe Ren](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=3t2j87YAAAAJ)\u003C\u002Fli> \u003Cli>[Zhongdao Wang](https:\u002F\u002Fzhongdao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ping Luo](http:\u002F\u002Fluoping.me\u002F)\u003C\u002Fli> \u003Cli>[Huchuan Lu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=D3nE0agAAAAJ)\u003C\u002Fli> \u003Cli>[Zhenguo Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=XboZC1AAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPixArt-alpha\u002FPixArt-sigma?style=social)](https:\u002F\u002Fgithub.com\u002FPixArt-alpha\u002FPixArt-sigma) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2403.04692), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.00426), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2401.05252)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002Frde6eaE5Ta)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FPixArt-alpha\u002FPixArt-alpha), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FPixArt-alpha\u002FPixArt-LCM)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpixart-alpha.github.io\u002FPixArt-sigma-project\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FPixArtSigma\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1jZ5UZXk7tcpTfVwnX33dDuefNMcnW9ME) | 07.11.2023 |\n| Zero123++ | Image-conditioned diffusion model for generating 3D-consistent multi-view images from a single input view | \u003Cul>\u003Cli>[Ruoxi Shi](https:\u002F\u002Frshi.top\u002F)\u003C\u002Fli> \u003Cli>[Hansheng Chen](https:\u002F\u002Flakonik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhuoyang Zhang](https:\u002F\u002Fgithub.com\u002Fzhuoyang20)\u003C\u002Fli> \u003Cli>[Minghua Liu](https:\u002F\u002Fcseweb.ucsd.edu\u002F~mil070\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chao Xu](https:\u002F\u002Fchaoxu.xyz\u002F)\u003C\u002Fli> \u003Cli>[Xinyue Wei](https:\u002F\u002Fsarahweiii.github.io\u002F)\u003C\u002Fli> \u003Cli>[Linghao Chen](https:\u002F\u002Footts.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chong Zeng](https:\u002F\u002Fwww.chong-zeng.com\u002F)\u003C\u002Fli> \u003Cli>[Hao Su](https:\u002F\u002Fcseweb.ucsd.edu\u002F~haosu\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FSUDO-AI-3D\u002Fzero123plus?style=social)](https:\u002F\u002Fgithub.com\u002FSUDO-AI-3D\u002Fzero123plus) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.15110)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOne-2-3-45\u002FOne-2-3-45), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcvlab-columbia\u002Fzero123)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fsudo-ai\u002Fzero123plus-demo-space), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fysharma\u002FZero123PlusDemo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fxthemadgenius.medium.com\u002Fzero123-your-guide-to-single-view-to-multi-view-3d-image-transformation-b4346b0e6615)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F17f4c6p\u002Fzero123_a_single_image_to_consistent_multiview\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FV9AR-81pAgk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1_5ECnTOosRuAsm2tUp0zvBG0DppL-F3V) | 26.10.2023 |\n| UniFormerV2 | Unified Transformer for Efficient Spatiotemporal Representation Learning | \u003Cul>\u003Cli>[Kunchang Li](https:\u002F\u002Fgithub.com\u002FAndy1621)\u003C\u002Fli> \u003Cli>[Yali Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=hD948dkAAAAJ)\u003C\u002Fli> \u003Cli>[Yinan He](https:\u002F\u002Fgithub.com\u002Fyinanhe)\u003C\u002Fli> \u003Cli>[Yizhuo Li](http:\u002F\u002Fliyizhuo.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yi Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Xm2M8UwAAAAJ)\u003C\u002Fli> \u003Cli>[Limin Wang](http:\u002F\u002Fwanglimin.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yu Qiao](http:\u002F\u002Fmmlab.siat.ac.cn\u002Fyuqiao\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cba6c3f73678.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.00157) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FOpenGVLab\u002FUniFormerV2?style=social)](https:\u002F\u002Fgithub.com\u002FOpenGVLab\u002FUniFormerV2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.09552)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Finnat\u002FUniFormerV2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FAndy1621\u002Funiformerv2_demo), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fpytorch-image-models\u002Fblob\u002Fmain\u002Ftimm\u002Fmodels\u002Fvision_transformer.py), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FSlowFast)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FAndy1621\u002Funiformerv2_demo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Faction-classification-on-activitynet?p=uniformerv2-spatiotemporal-learning-by-arming), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Faction-recognition-on-hacs?p=uniformerv2-spatiotemporal-learning-by-arming), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Faction-classification-on-moments-in-time?p=uniformerv2-spatiotemporal-learning-by-arming), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Faction-recognition-in-videos-on-something-1?p=uniformerv2-spatiotemporal-learning-by-arming), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Faction-classification-on-kinetics-700?p=uniformerv2-spatiotemporal-learning-by-arming)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Z6RzLcno_eLGD_E96mlWoyGieGbKxPQr) | 20.10.2023 |\n| Show-1 | Hybrid model, dubbed as Show-1, which marries pixel-based and latent-based VDMs for text-to-video generation | \u003Cul>\u003Cli>[David Junhao Zhang](https:\u002F\u002Fjunhaozhang98.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jay Zhangjie Wu](https:\u002F\u002Fzhangjiewu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiawei Liu](https:\u002F\u002Fjia-wei-liu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Rui Zhao](https:\u002F\u002Fruizhaocv.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lingmin Ran](https:\u002F\u002Fsiacorplab.nus.edu.sg\u002Fpeople\u002Fran-lingmin\u002F)\u003C\u002Fli> \u003Cli>[Yuchao Gu](https:\u002F\u002Fycgu.site\u002F)\u003C\u002Fli> \u003Cli>[Difei Gao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=No9OsocAAAAJ)\u003C\u002Fli> \u003Cli>[Mike Zheng Shou](https:\u002F\u002Fsites.google.com\u002Fview\u002Fshowlab\u002Fhome)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fshowlab\u002FShow-1?style=social)](https:\u002F\u002Fgithub.com\u002Fshowlab\u002FShow-1) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.15818)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fshowlab\u002Fshow-1-base), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fshowlab\u002Fshow-1-interpolation), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fshowlab\u002Fshow-1-sr1), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fshowlab\u002Fshow-1-sr2), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdamo-vilab\u002Fmodelscope-damo-text-to-video-synthesis), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fcerspense\u002Fzeroscope_v2_576w)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fshowlab.github.io\u002FShow-1\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002FShow-1-colab\u002Fblob\u002Fmain\u002FShow_1_steps_colab.ipynb) | 15.10.2023 |\n| DA-CLIP | Degradation-aware vision-language model to better transfer pretrained vision-language models to low-level vision tasks as a universal framework for image restoration | \u003Cul>\u003Cli>[Ziwei Luo](https:\u002F\u002Falgolzw.github.io\u002F)\u003C\u002Fli> \u003Cli>[Fredrik Gustafsson](http:\u002F\u002Fwww.fregu856.com\u002F)\u003C\u002Fli> \u003Cli>[Zheng Zhao](https:\u002F\u002Fzz.zabemon.com\u002F)\u003C\u002Fli> \u003Cli>[Jens Sjölund](https:\u002F\u002Fgithub.com\u002Fjsjol)\u003C\u002Fli> \u003Cli>[Thomas Schön](https:\u002F\u002Fuser.it.uu.se\u002F~thosc112\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAlgolzw\u002Fdaclip-uir?style=social)](https:\u002F\u002Fgithub.com\u002FAlgolzw\u002Fdaclip-uir) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2310.01018)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAlgolzw\u002Fimage-restoration-sde)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fweblzw\u002Fdaclip-uir-ViT-B-32-irsde)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Falgolzw.github.io\u002Fdaclip-uir\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002Fdaclip-uir-colab\u002Fblob\u002Fmain\u002Fdaclip_uir_gradio_colab.ipynb) | 11.10.2023 |\n| SadTalker | Generates 3D motion coefficients of the 3DMM from audio and implicitly modulates a novel 3D-aware face render for talking head generation | \u003Cul>\u003Cli>[Wenxuan Zhang](https:\u002F\u002Fgithub.com\u002FWinfredy)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002Facademic\u002F)\u003C\u002Fli> \u003Cli>[Xuan Wang](https:\u002F\u002Fxuanwangvc.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yong Zhang](https:\u002F\u002Fyzhang2016.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xi Shen](https:\u002F\u002Fxishen0220.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yu Guo](https:\u002F\u002Fyuguo-xjtu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli> \u003Cli>[Fei Wang](http:\u002F\u002Fgr.xjtu.edu.cn\u002Fzh\u002Fweb\u002Ffeynmanw)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_125d6c68e8af.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52729.2023.00836) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FOpenTalker\u002FSadTalker?style=social)](https:\u002F\u002Fgithub.com\u002FOpenTalker\u002FSadTalker) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.12194)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FrrayYqZ4tf)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOpenTalker\u002FDPE), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhanglonghao1992\u002FOne-Shot_Free-View_Neural_Talking_Head_Synthesis), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FRenYurui\u002FPIRender), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FDeep3DFaceReconstruction), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002Ffacexlib), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZz-ww\u002FSadTalker-Video-Lip-Sync), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFeiiYin\u002FSPI)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsadtalker.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAoIzJWnQw1M), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfDgQcDL-qOc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBkSnM9cxkcM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F7u0FYVPQ5rc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FOpenTalker\u002FSadTalker\u002Fblob\u002Fmain\u002Fquick_demo.ipynb) | 10.10.2023 |\n| Musika | Music generation system that can be trained on hundreds of hours of music using a single consumer GPU, and that allows for much faster than real-time generation of music of arbitrary length on a consumer CPU | \u003Cul>\u003Cli>[Marco Pasini](https:\u002F\u002Fgithub.com\u002Fmarcoppasini)\u003C\u002Fli> \u003Cli>[Jan Schlüter](https:\u002F\u002Fwww.ofai.at\u002F~jan.schlueter\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmarcoppasini\u002Fmusika?style=social)](https:\u002F\u002Fgithub.com\u002Fmarcoppasini\u002Fmusika) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.08706), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.08526)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fmagenta.tensorflow.org\u002Fdatasets\u002Fmaestro)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhendriks73\u002Ftempo-cnn), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCPJKU\u002Fmadmom)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fmarcop\u002Fmusika)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmarcoppasini.github.io\u002Fmusika)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQBl8y2Z_i7Y), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0l7OSM-bFvc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1PowSw3doBURwLE-OTCiWkO8HVbS5paRb) | 09.10.2023 |\n| YOLOv6 | Single-stage object detection framework dedicated to industrial applications | \u003Cul>\u003Cli>[Kaiheng Weng](https:\u002F\u002Fgithub.com\u002FkhwengXU)\u003C\u002Fli> \u003Cli>[Meng Cheng](https:\u002F\u002Fgithub.com\u002FMTChengMeng)\u003C\u002Fli> \u003Cli>[Yiduo Li](https:\u002F\u002Fgithub.com\u002Fyili123123)\u003C\u002Fli> \u003Cli>[Xiangxiang Chu](https:\u002F\u002Fscholar.google.com\u002Fcitations?&user=jn21pUsAAAAJ)\u003C\u002Fli> \u003Cli>[Xiaolin Wei](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=s5b7lU4AAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmeituan\u002FYOLOv6?style=social)](https:\u002F\u002Fgithub.com\u002Fmeituan\u002FYOLOv6) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.02976), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.05586)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Flearnopencv.com\u002Fyolov6-object-detection\u002F)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcocodataset.org\u002F#download)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fyolov6-docs.readthedocs.io\u002Fzh_CN\u002Flatest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFeiGeChuanShu\u002Fncnn-android-yolov6), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDefTruth\u002Flite.ai.toolkit\u002Fblob\u002Fmain\u002Flite\u002Fort\u002Fcv\u002Fyolov6.cpp), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FLinaom1214\u002FTensorRT-For-YOLO-Series), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhiqwang\u002Fyolov5-rt-stack\u002Ftree\u002Fmain\u002Fdeployment\u002Ftensorrt-yolov6)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3OpwcGU7VvE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGJ0lVOE3a7c), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3hqkbqJ5ag8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfFCWrMFH2UY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmeituan\u002FYOLOv6\u002Fblob\u002Fmaster\u002Fturtorial.ipynb) | 08.10.2023 |\n| DreamGaussian | Algorithm to convert 3D Gaussians into textured meshes and apply a fine-tuning stage to refine the details | \u003Cul>\u003Cli>[Jiaxiang Tang](https:\u002F\u002Fme.kiui.moe\u002F)\u003C\u002Fli> \u003Cli>[Jiawei Ren](https:\u002F\u002Fjiawei-ren.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hang Zhou](https:\u002F\u002Fhangz-nju-cuhk.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gang Zeng](http:\u002F\u002Fwww.cis.pku.edu.cn\u002Finfo\u002F1177\u002F1378.htm)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdreamgaussian\u002Fdreamgaussian?style=social)](https:\u002F\u002Fgithub.com\u002Fdreamgaussian\u002Fdreamgaussian) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2309.16653)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgraphdeco-inria\u002Fdiff-gaussian-rasterization), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fnvdiffrast), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhoffstadt\u002FDearPyGui)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fdreamgaussian.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1sLpYmmLS209-e5eHgcuqdryFRRO6ZhFS) | 04.10.2023 |\n| ICON | Given a set of images, method estimates a detailed 3D surface from each image and then combines these into an animatable avatar | \u003Cul>\u003Cli>[Yuliang Xiu](https:\u002F\u002Fxiuyuliang.cn\u002F)\u003C\u002Fli> \u003Cli>[Jinlong Yang](https:\u002F\u002Fis.mpg.de\u002F~jyang)\u003C\u002Fli> \u003Cli>[Dimitrios Tzionas](https:\u002F\u002Fps.is.mpg.de\u002F~dtzionas)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2bcf751092ab.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01294) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyuliangxiu\u002Ficon?style=social)](https:\u002F\u002Fgithub.com\u002Fyuliangxiu\u002Ficon) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.09127)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FKeypointNeRF), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYadiraF\u002FPIXIE), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYuliangXiu\u002Fbvh-distance-queries), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FProject-Splinter\u002FMonoPortDataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FZhengZerong\u002FPaMIR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FProject-Splinter\u002FMonoPort), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fshunsukesaito\u002FSCANimate), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Faistplusplus_api)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FYuliang\u002FICON)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ficon.is.tue.mpg.de\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhZd6AYin2DE)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1-AWeWhPvCTBX0KfMtgtMk10uPU05ihoA) | 31.08.2023 |\n| DINOv2 | Produce high-performance visual features that can be directly employed with classifiers as simple as linear layers on a variety of computer vision tasks; these visual features are robust and perform well across domains without any requirement for fine-tuning | \u003Cul>\u003Cli>[Maxime Oquab](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5vteYV8AAAAJ)\u003C\u002Fli> \u003Cli>[Timothée Darcet](https:\u002F\u002Fgithub.com\u002FTimDarcet)\u003C\u002Fli> \u003Cli>[Théo Moutakanni](https:\u002F\u002Fgithub.com\u002FTheoMoutakanni)\u003C\u002Fli> \u003Cli>[Huy Vo](https:\u002F\u002Fhuyvvo.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Marc Szafraniec](https:\u002F\u002Fgithub.com\u002FMarcSzafraniec\u002F)\u003C\u002Fli> \u003Cli>[Vasil Khalidov](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=tjazz3AAAAAJ)\u003C\u002Fli> \u003Cli>[Pierre Fernandez](https:\u002F\u002Fpierrefdz.github.io\u002F)\u003C\u002Fli> \u003Cli>[Daniel Haziza](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2eSKdFMAAAAJ)\u003C\u002Fli> \u003Cli>[Francisco Massa](https:\u002F\u002Fgithub.com\u002Ffmassa)\u003C\u002Fli> \u003Cli>[Alaaeldin El-Nouby](https:\u002F\u002Faelnouby.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mahmoud Assran](http:\u002F\u002Fwww.midoassran.ca\u002F)\u003C\u002Fli> \u003Cli>[Nicolas Ballas](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=euUV4iUAAAAJ)\u003C\u002Fli> \u003Cli>[Wojciech Galuba](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jyaTX64AAAAJ)\u003C\u002Fli> \u003Cli>[Russell Howes](http:\u002F\u002Fwww.russellhowes.net\u002F)\u003C\u002Fli> \u003Cli>[Po-Yao Huang](https:\u002F\u002Fberniebear.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shang-Wen Li](https:\u002F\u002Fswdanielli.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ishan Misra](http:\u002F\u002Fimisra.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michael Rabbat](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=cMPKe9UAAAAJ)\u003C\u002Fli> \u003Cli>[Vasu Sharma](https:\u002F\u002Fvasusharma.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gabriel Synnaeve](https:\u002F\u002Fsyhw.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hu Xu](https:\u002F\u002Fhowardhsu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hervé Jégou](https:\u002F\u002Fgithub.com\u002Fjegou)\u003C\u002Fli> \u003Cli>[Julien Mairal](http:\u002F\u002Fthoth.inrialpes.fr\u002Fpeople\u002Fmairal\u002F)\u003C\u002Fli> \u003Cli>[Patrick Labatut](https:\u002F\u002Fgithub.com\u002Fpatricklabatut)\u003C\u002Fli> \u003Cli>[Armand Joulin](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=kRJkDakAAAAJ)\u003C\u002Fli> \u003Cli>[Piotr Bojanowski](https:\u002F\u002Fgithub.com\u002Fpiotr-bojanowski)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fdinov2?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdinov2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.07193)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fai.facebook.com\u002Fblog\u002Fdino-v2-computer-vision-self-supervised-learning\u002F)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fdinov2.metademolab.com\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmain\u002Fmodel_doc\u002Fdinov2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fpurnasaigudikandula.medium.com\u002Fdinov2-image-classification-visualization-and-paper-review-745bee52c826), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Ftowardsdatascience.com\u002Fmeta-ais-another-revolutionary-large-scale-model-dinov2-for-image-feature-extraction-1114b287eadd)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FcsEgtSh7jV4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FKSZiJ4k28b4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRZEkdOc3szU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fdinov2\u002Fblob\u002Fmain\u002Fnotebooks\u002Fsemantic_segmentation.ipynb) | 31.08.2023 |\n| StyleGAN 3 | Alias-Free Generative Adversarial Networks | \u003Cul>\u003Cli>[Tero Karras](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Ftero-karras)\u003C\u002Fli> \u003Cli>[Miika Aittala](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002FMiika-Aittala)\u003C\u002Fli> \u003Cli>[Samuli Laine](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002FSamuli-Laine)\u003C\u002Fli> \u003Cli>[Erik Härkönen](https:\u002F\u002Fgithub.com\u002Fharskish)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Janne Hellsten](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002FJanne-Hellsten)\u003C\u002Fli> \u003Cli>[Jaakko Lehtinen](https:\u002F\u002Fusers.aalto.fi\u002F~lehtinj7\u002F)\u003C\u002Fli> \u003Cli>[Timo Aila](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Ftimo-aila)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNVlabs\u002Fstylegan3?style=social)](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan3) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.12423), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.08500), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.01401), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.06991), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.04948), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1606.03498)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan3-detector), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fmetfaces-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2021\u002Fhash\u002F076ccd93ad68be51f23707988e934906-Abstract.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnvlabs.github.io\u002Fstylegan3)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1BXNHZBai-pXtP-ncliouXo_kUiG1Pq7M) | 13.08.2023 |\n| FateZero | Zero-shot text-based editing method on real-world videos without per-prompt training or use-specific mask | \u003Cul>\u003Cli>[Chenyang Qi](https:\u002F\u002Fchenyangqiqi.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002Facademic\u002F)\u003C\u002Fli> \u003Cli>[Yong Zhang](https:\u002F\u002Fyzhang2016.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chenyang Lei](https:\u002F\u002Fchenyanglei.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli> \u003Cli>[Qifeng Chen](https:\u002F\u002Fcqf.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a4c0093cdc32.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.01460) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FChenyangQiQi\u002FFateZero?style=social)](https:\u002F\u002Fgithub.com\u002FChenyangQiQi\u002FFateZero) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.09535)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbryandlee\u002FTune-A-Video), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fprompt-to-prompt)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fchenyangqi\u002FFateZero), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fchenyangqi\u002Fjeep_tuned_200)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ffate-zero-edit.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002F11uzioo\u002Fr_fatezero_fusing_attentions_for_zeroshot\u002F)\u003C\u002Fli>\u003Cli>[video](https:\u002F\u002Fhkustconnect-my.sharepoint.com\u002Fpersonal\u002Fcqiaa_connect_ust_hk\u002F_layouts\u002F15\u002Fstream.aspx?id=%2Fpersonal%2Fcqiaa%5Fconnect%5Fust%5Fhk%2FDocuments%2Fdiffusion%2Fweb%5Fvideo%2Emp4&ga=1&referrer=StreamWebApp%2EWeb&referrerScenario=AddressBarCopied%2Eview%2E9b85614a%2D5af9%2D4485%2Dbcb1%2Db39f90e8d381)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FChenyangQiQi\u002FFateZero\u002Fblob\u002Fmain\u002Fcolab_fatezero.ipynb) | 13.08.2023 |\n| Big GAN | Large Scale GAN Training for High Fidelity Natural Image Synthesis | \u003Cul>\u003Cli>[Andrew Brock](https:\u002F\u002Fgithub.com\u002Fajbrock)\u003C\u002Fli> \u003Cli>[Jeff Donahue](https:\u002F\u002Fjeffdonahue.com\u002F)\u003C\u002Fli> \u003Cli>[Karen Simonyan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=L7lMQkQAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.11096)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fhub\u002Fblob\u002Fmaster\u002Fexamples\u002Fcolab\u002Fbiggan_generation_with_tf_hub.ipynb) | 03.08.2023 |\n| LaMa | Resolution-robust Large Mask Inpainting with Fourier Convolutions | \u003Cul>\u003Cli>[Roman Suvorov](https:\u002F\u002Fgithub.com\u002Fwindj007)\u003C\u002Fli> \u003Cli>[Elizaveta Logacheva](https:\u002F\u002Fgithub.com\u002Felimohl)\u003C\u002Fli> \u003Cli>[Anton Mashikhin](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fheyt0ny\u002F)\u003C\u002Fli> \u003Cli>[Anastasia Remizova](https:\u002F\u002Fgithub.com\u002Ffeathernox)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Arsenii Ashukha](https:\u002F\u002Fashukha.com\u002F)\u003C\u002Fli> \u003Cli>[Aleksei Silvestrov](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002F%D0%B0%D0%BB%D0%B5%D0%BA%D1%81%D0%B5%D0%B9-%D1%81%D0%B8%D0%BB%D1%8C%D0%B2%D0%B5%D1%81%D1%82%D1%80%D0%BE%D0%B2-141b99b6\u002F)\u003C\u002Fli> \u003Cli>[Naejin Kong](https:\u002F\u002Fgithub.com\u002Fnaejin-kong)\u003C\u002Fli> \u003Cli>[Harshith Goka](https:\u002F\u002Fgithub.com\u002Fh9399-goka)\u003C\u002Fli> \u003Cli>[Kiwoong Park](https:\u002F\u002Fgithub.com\u002Fkyoong-park)\u003C\u002Fli> \u003Cli>[Victor Lempitsky](http:\u002F\u002Fsites.skoltech.ru\u002Fcompvision\u002Fmembers\u002Fvilem\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_f3f60f020173.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FWACV51458.2022.00323) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsaic-mdal\u002Flama?style=social)](https:\u002F\u002Fgithub.com\u002Fsaic-mdal\u002Flama) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.07161)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fandy971022\u002Fauto-lama), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frichzhang\u002FPerceptualSimilarity), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FPo-Hsun-Su\u002Fpytorch-ssim), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmseitzer\u002Fpytorch-fid)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsaic-mdal.github.io\u002Flama-project\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsaic-mdal\u002Flama\u002Fblob\u002Fmaster\u002Fcolab\u002FLaMa_inpainting.ipynb) | 02.08.2023 |\n| MakeItTalk | A method that generates expressive talking-head videos from a single facial image with audio as the only input | \u003Cul>\u003Cli>[Yang Zhou](https:\u002F\u002Fpeople.umass.edu\u002F~yangzhou\u002F)\u003C\u002Fli> \u003Cli>[Xintong Han](http:\u002F\u002Fusers.umiacs.umd.edu\u002F~xintong\u002F)\u003C\u002Fli> \u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli> \u003Cli>[Jose Echevarria](http:\u002F\u002Fwww.jiechevarria.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Evangelos Kalogerakis](https:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002F)\u003C\u002Fli> \u003Cli>[Dingzeyu Li](https:\u002F\u002Fdingzeyu.li\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_84498a2017f8.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3414685.3417774) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyzhou359\u002FMakeItTalk?style=social)](https:\u002F\u002Fgithub.com\u002Fyzhou359\u002FMakeItTalk) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.12992)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fdrive.google.com\u002Fdrive\u002Ffolders\u002F1EwuAy3j1b9Zc1MsidUfxG_pJGc_cV60O)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpeople.umass.edu\u002F~yangzhou\u002FMakeItTalk\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=vUMGKASgbf8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fiboyles\u002Fmakeittalknow\u002Fblob\u002Fmain\u002Fworking_quick_demo_of_makeittalk_07_2023.ipynb) | 27.07.2023 |\n| HiDT | A generative image-to-image model and a new upsampling scheme that allows to apply image translation at high resolution | \u003Cul>\u003Cli>[Denis Korzhenkov](https:\u002F\u002Fgithub.com\u002Fdenkorzh)\u003C\u002Fli> \u003Cli>[Gleb Sterkin](https:\u002F\u002Fgithub.com\u002Fbelkakari)\u003C\u002Fli> \u003Cli>[Sergey Nikolenko](https:\u002F\u002Flogic.pdmi.ras.ru\u002F~sergey\u002F)\u003C\u002Fli> \u003Cli>[Victor Lempitsky](http:\u002F\u002Fsites.skoltech.ru\u002Fcompvision\u002Fmembers\u002Fvilem\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_8388d5acf22d.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00751) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsaic-mdal\u002FHiDT?style=social)](https:\u002F\u002Fgithub.com\u002Fsaic-mdal\u002FHiDT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.08791)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsaic-mdal.github.io\u002FHiDT\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLuvGzlEQXT1KQuKrfBBEWh2f3PToxyeM5), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=EWKAgwgqXB4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsaic-mdal\u002Fhidt\u002Fblob\u002Fmaster\u002Fnotebooks\u002FHighResolutionDaytimeTranslation.ipynb) | 24.07.2023 |\n| AWQ | Activation-aware Weight Quantization, a hardware-friendly approach for LLM low-bit weight-only quantization | \u003Cul>\u003Cli>[Ji Lin](http:\u002F\u002Flinji.me\u002F)\u003C\u002Fli> \u003Cli>[Jiaming Tang](http:\u002F\u002Fjiamingtang.me\u002F)\u003C\u002Fli> \u003Cli>[Haotian Tang](https:\u002F\u002Fkentang.net\u002F)\u003C\u002Fli> \u003Cli>[Shang Yang](https:\u002F\u002Fgithub.com\u002Fys-2020)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Wei-Ming Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=6xFvyJwAAAAJ)\u003C\u002Fli> \u003Cli>[Wei-Chen Wang](https:\u002F\u002Fweichenwang.me\u002F)\u003C\u002Fli> \u003Cli>[Guangxuan Xiao](https:\u002F\u002Fguangxuanx.com\u002F)\u003C\u002Fli> \u003Cli>[Xingyu Dang](https:\u002F\u002Fgithub.com\u002Fdangxingyu)\u003C\u002Fli> \u003Cli>[Chuang Gan](https:\u002F\u002Fgithub.com\u002Fchuangg)\u003C\u002Fli> \u003Cli>[Song Han](https:\u002F\u002Fsonghan.mit.edu\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_260e9c4ff296.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3714983.3714987) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmit-han-lab\u002Fllm-awq?style=social)](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fllm-awq) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.00978), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2312.07533), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.17323)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fvila.hanlab.ai\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvllm-project\u002Fvllm\u002Fblob\u002Fmain\u002Fvllm\u002Fmodel_executor\u002Flayers\u002Fquantization\u002Fawq.py), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FTensorRT-LLM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Ftext-generation-inference\u002Fpull\u002F1054), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcasper-hansen\u002FAutoAWQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002FVILA), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fsmoothquant)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets\u002Fmit-han-lab\u002Fawq-model-zoo)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fbyte-sized-ai\u002Fvllm-quantization-awq-activation-aware-weight-quantization-for-llm-compression-and-35894ffd6a9b)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fhanlab.mit.edu\u002Fprojects\u002Fawq)\u003C\u002Fli>\u003Cli>[slides](https:\u002F\u002Fwww.dropbox.com\u002Fscl\u002Ffi\u002Fdtnp6h6y1mnp7g036axu6\u002FAWQ-slide.pdf)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdcINVsqxQgQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3dYLj9vjfA0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FOMkyocVyEpQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmit-han-lab\u002Fllm-awq\u002Fblob\u002Fmaster\u002Fexamples\u002Fchat_demo.ipynb) | 24.07.2023 |\n| Once-for-All | Train a once-for-all network that supports diverse architectural settings by decoupling training and search, to reduce the cost | \u003Cul>\u003Cli>[Han Cai](https:\u002F\u002Fhan-cai.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chuang Gan](https:\u002F\u002Fpeople.csail.mit.edu\u002Fganchuang\u002F)\u003C\u002Fli> \u003Cli>[Tianzhe Wang](https:\u002F\u002Fsites.google.com\u002Fview\u002Ftianzhe-wang\u002Fhome)\u003C\u002Fli> \u003Cli>[Zhekai Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=aYh01A4AAAA)\u003C\u002Fli> \u003Cli>[Song Han](https:\u002F\u002Fhanlab.mit.edu\u002Fsonghan)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_3ec06e4edff8.png)](https:\u002F\u002Fdoi.org\u002F10.48550\u002FarXiv.1908.09791) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmit-han-lab\u002Fonce-for-all?style=social)](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fonce-for-all) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.00332), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.03494), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.08886)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsony\u002Fnnabla-nas), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fanalogdevicesinc\u002Fai8x-training)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fpytorch.org\u002Fhub\u002Fpytorch_vision_once_for_all)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fofa\u002F)\u003C\u002Fli>\u003Cli>[slides](https:\u002F\u002Ffile.lzhu.me\u002Fprojects\u002FOnceForAll\u002FOFA%20Slides.pdf)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fa_OeT8MXzWI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fwrsid5tvuSM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfptQ_eJ3Uc0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjsyHqDX5cU8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fg3ujV7q0wZk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FLgt2kg1_pTI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FFVfh2vw4RG0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FWHQWlIKdwsk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmit-han-lab\u002Fonce-for-all\u002Fblob\u002Fmaster\u002Ftutorial\u002Fofa.ipynb) | 19.07.2023 |\n| Recognize Anything & Tag2Text | Vision language pre-training framework, which introduces image tagging into vision-language models to guide the learning of visual-linguistic features | \u003Cul>\u003Cli>[Xinyu Huang](https:\u002F\u002Fxinyu1205.github.io\u002F)\u003C\u002Fli> \u003Cli>[Youcai Zhang](https:\u002F\u002Fgithub.com\u002FColer1994)\u003C\u002Fli> \u003Cli>[Jinyu Ma](https:\u002F\u002Fgithub.com\u002Fmajinyu666)\u003C\u002Fli> \u003Cli>[Zhaoyang Li](https:\u002F\u002Fgithub.com\u002FZhaoyangLi-nju)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yanchun Xie](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=T0xk9-wAAAAJ)\u003C\u002Fli> \u003Cli>[Yuzhuo Qin](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5ZG65AkAAAAJ)\u003C\u002Fli> \u003Cli>[Tong Luo](https:\u002F\u002Fieeexplore.ieee.org\u002Fauthor\u002F37089387319)\u003C\u002Fli> \u003Cli>[Yaqian Li](https:\u002F\u002Fopenreview.net\u002Fprofile?id=~Yaqian_Li1)\u003C\u002Fli> \u003Cli>[Yandong Guo](http:\u002F\u002Fwww.lsl.zone\u002F)\u003C\u002Fli> \u003Cli>[Yandong Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=fWDoWsQAAAAJ)\u003C\u002Fli> \u003Cli>[Lei Zhang](https:\u002F\u002Fwww.leizhang.org\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fxinyu1205\u002Frecognize-anything?style=social)](https:\u002F\u002Fgithub.com\u002Fxinyu1205\u002Frecognize-anything) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.03514), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.05657)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOpenGVLab\u002FAsk-Anything), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fpositive666\u002FPrompt-Can-Anything)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fartgor.medium.com\u002Fpaper-review-recognize-anything-a-strong-image-tagging-model-9e5e1c6dd0af)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Frecognize-anything.github.io\u002F), [project](https:\u002F\u002Frecognize-anything.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmhd-medfa\u002Frecognize-anything\u002Fblob\u002Fmain\u002Frecognize_anything_demo.ipynb) | 09.07.2023 |\n| Thin-Plate Spline Motion Model | End-to-end unsupervised motion transfer framework | \u003Cul>\u003Cli>[Jian Zhao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=OKm5CQYAAAAJ)\u003C\u002Fli> \u003Cli>[Hui Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=w3mzCiwAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_60204a493e21.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00364) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyoyo-nb\u002FThin-Plate-Spline-Motion-Model?style=social)](https:\u002F\u002Fgithub.com\u002Fyoyo-nb\u002FThin-Plate-Spline-Motion-Model) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.14367)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Fmonkey-net), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Fvideo-preprocessing), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Fpose-evaluation), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTalkUHulk\u002FImage-Animation-Turbo-Boost)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FCVPR\u002FImage-Animation-using-Thin-Plate-Spline-Motion-Model)\u003C\u002Fli>\u003Cli>[supp](https:\u002F\u002Fcloud.tsinghua.edu.cn\u002Ff\u002Ff7b8573bb5b04583949f\u002F?dl=1)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1DREfdpnaBhqISg0fuQlAAIwyGVn1loH_) | 07.07.2023 |\n| MobileSAM | Towards Lightweight SAM for Mobile Applications | \u003Cul>\u003Cli>[Chaoning Zhang](https:\u002F\u002Fgithub.com\u002FChaoningZhang)\u003C\u002Fli> \u003Cli>[Dongshen Han](https:\u002F\u002Fgithub.com\u002Fdongshenhan)\u003C\u002Fli> \u003Cli>[Yu Qiao](https:\u002F\u002Fgithub.com\u002Fqiaoyu1002)\u003C\u002Fli> \u003Cli>[Jung Uk Kim](https:\u002F\u002Fvisualai.khu.ac.kr\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Sung-Ho Bae](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=EULut5oAAAAJ)\u003C\u002Fli> \u003Cli>[Seungkyu Lee](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=3Pf6C6cAAAAJ)\u003C\u002Fli> \u003Cli>[Choong Seon Hong](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=oKANWloAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FChaoningZhang\u002FMobileSAM?style=social)](https:\u002F\u002Fgithub.com\u002FChaoningZhang\u002FMobileSAM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2306.14289)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjolibrain\u002FjoliGEN), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fakbartus\u002FMobileSAM-in-the-Browser), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fqiaoyu1002\u002FInpaint-Anything), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fqiaoyu1002\u002FPersonalize-SAM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJumpat\u002FSegmentAnythingin3D), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvietanhdev\u002Fanylabeling), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fwangsssky\u002FSonarSAM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcontinue-revolution\u002Fsd-webui-segment-anything)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002F_akhaliq\u002Fstatus\u002F1674410573075718145)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FeTEfq_kWabQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FChaoningZhang\u002FMobileSAM\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fpredictor_example.ipynb) | 30.06.2023 |\n| T5X | Modular, composable, research-friendly framework for high-performance, configurable, self-service training, evaluation, and inference of sequence models at many scales | \u003Cul>\u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Hyung Won Chung](https:\u002F\u002Fgithub.com\u002Fhwchung27)\u003C\u002Fli> \u003Cli>[Anselm Levskaya](https:\u002F\u002Fanselmlevskaya.com\u002F)\u003C\u002Fli> \u003Cli>[Gaurav Mishra](https:\u002F\u002Fresearch.google\u002Fpeople\u002FGauravMishra\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[James Bradbury](https:\u002F\u002Fgithub.com\u002Fjekbradbury)\u003C\u002Fli> \u003Cli>[Daniel Andor](https:\u002F\u002Fgithub.com\u002Fandorardo)\u003C\u002Fli> \u003Cli>[Sharan Narang](https:\u002F\u002Fgithub.com\u002Fsharannarang)\u003C\u002Fli> \u003Cli>[Brian Lester](https:\u002F\u002Fblester125.com\u002F)\u003C\u002Fli> \u003Cli>[Colin Gaffney](https:\u002F\u002Fgithub.com\u002Fcpgaffney1)\u003C\u002Fli> \u003Cli>[Afroz Mohiuddin](https:\u002F\u002Fgithub.com\u002Fafrozenator)\u003C\u002Fli> \u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli> \u003Cli>[Aitor Lewkowycz](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Yum1ah0AAAAJ)\u003C\u002Fli> \u003Cli>[Alex Salcianu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=HSrT1wsAAAAJ)\u003C\u002Fli> \u003Cli>[Marc van Zee](https:\u002F\u002Fgithub.com\u002Fmarcvanzee)\u003C\u002Fli> \u003Cli>[Jacob Austin](https:\u002F\u002Fjacobaustin123.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sebastian Goodman](https:\u002F\u002Fgithub.com\u002F0x0539)\u003C\u002Fli> \u003Cli>[Livio Baldini Soares](https:\u002F\u002Fliviosoares.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haitang Hu](https:\u002F\u002Fhthu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sasha Tsvyashchenko](https:\u002F\u002Fendl.ch\u002F)\u003C\u002Fli> \u003Cli>[Aakanksha Chowdhery](http:\u002F\u002Fwww.achowdhery.com\u002F)\u003C\u002Fli> \u003Cli>[Jasmijn Bastings](https:\u002F\u002Fjasmijn.ninja\u002F)\u003C\u002Fli> \u003Cli>[Jannis Bulian](http:\u002F\u002Fbulian.org\u002F)\u003C\u002Fli> \u003Cli>[Xavier Garcia](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Y2Hio6MAAAAJ)\u003C\u002Fli> \u003Cli>[Jianmo Ni](https:\u002F\u002Fnijianmo.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kathleen Kenealy](https:\u002F\u002Fscholar.google.com\u002Fcitations?&user=HgRBC5gAAAAJ)\u003C\u002Fli> \u003Cli>[Jonathan Clark](http:\u002F\u002Fwww.cs.cmu.edu\u002F~jhclark\u002F)\u003C\u002Fli> \u003Cli>[Dan Garrette](http:\u002F\u002Fwww.dhgarrette.com\u002F)\u003C\u002Fli> \u003Cli>[James Lee-Thorp](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=qsPv098AAAAJ)\u003C\u002Fli> \u003Cli>[Colin Raffel](https:\u002F\u002Fcolinraffel.com\u002F)\u003C\u002Fli> \u003Cli>[Noam Shazeer](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=wsGvgA8AAAAJ)\u003C\u002Fli> \u003Cli>[Marvin Ritter](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=arcf5FgAAAAJ)\u003C\u002Fli> \u003Cli>[Maarten Bosma](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=wkeFQPgAAAAJ)\u003C\u002Fli> \u003Cli>[Alexandre Passos](https:\u002F\u002Fwww.ic.unicamp.br\u002F~tachard\u002F)\u003C\u002Fli> \u003Cli>[Jeremy Maitin-Shepard](https:\u002F\u002Fresearch.google\u002Fpeople\u002FJeremyMaitinShepard\u002F)\u003C\u002Fli> \u003Cli>[Noah Fiedel](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=XWpV9DsAAAAJ)\u003C\u002Fli> \u003Cli>[Brennan Saeta](https:\u002F\u002Fgithub.com\u002Fsaeta)\u003C\u002Fli> \u003Cli>[Ryan Sepassi](https:\u002F\u002Fryansepassi.com\u002F)\u003C\u002Fli> \u003Cli>[Alexander Spiridonov](https:\u002F\u002Fresearch.google\u002Fpeople\u002FAlexanderSpiridonov\u002F)\u003C\u002Fli> \u003Cli>[Joshua Newlan](https:\u002F\u002Fgithub.com\u002Fjoshnewlan)\u003C\u002Fli> \u003Cli>[Andrea Gesmundo](https:\u002F\u002Fgithub.com\u002Fagesmundo)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Ft5x?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Ft5x) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.17189), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.10683)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Ft5x.readthedocs.io\u002Fen\u002Flatest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fmesh), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fserving)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fdatasets\u002Fcatalog\u002Fwmt_t2t_translate), [\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fguide\u002Fdata), [\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Ftensorboard)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Ft5x\u002Fblob\u002Fmain\u002Ft5x\u002Fnotebooks\u002Fintroduction.ipynb) | 27.06.2023 |\n| CodeTalker | Cast speech-driven facial animation as a code query task in a finite proxy space of the learned codebook, which effectively promotes the vividness of the generated motions by reducing the cross-modal mapping uncertainty | \u003Cul>\u003Cli>[Jinbo Xing](https:\u002F\u002Fdoubiiu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Menghan Xia](https:\u002F\u002Fmenghanxia.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuechen Zhang](https:\u002F\u002Fjulianjuaner.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002Facademic\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jue Wang](https:\u002F\u002Fjuewang725.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tien-Tsin Wong](https:\u002F\u002Fttwong12.github.io\u002Fmyself.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_de3910b4cb3f.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52729.2023.01229) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FDoubiiu\u002FCodeTalker?style=social)](https:\u002F\u002Fgithub.com\u002FDoubiiu\u002FCodeTalker) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2301.02379), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.09797)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMPI-IS\u002Fmesh), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTimoBolkart\u002Fvoca\u002Ftree\u002Fmaster\u002Ftemplate), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEvelynFan\u002FFaceFormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FRenYurui\u002FPIRender), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOpenTalker\u002FStyleHEAT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMeta-Portrait\u002FMetaPortrait)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fdoubiiu.github.io\u002Fprojects\u002Fcodetalker\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJ2RngmuYrG4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FDoubiiu\u002FCodeTalker\u002Fblob\u002Fmain\u002Fdemo.ipynb) | 16.06.2023 |\n| Gen-L-Video | Extending off-the-shelf short video diffusion models for generating and editing videos comprising hundreds of frames with diverse semantic segments without introducing additional training, all while preserving content consistency | \u003Cul>\u003Cli>[Fu-Yun Wang](https:\u002F\u002Fg-u-n.github.io\u002F)\u003C\u002Fli> \u003Cli>[Wenshuo Chen](https:\u002F\u002Fgithub.com\u002Fwinshot-thu)\u003C\u002Fli> \u003Cli>[Guanglu Song](https:\u002F\u002Fsongguanglu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Han-Jia Ye](https:\u002F\u002Fgithub.com\u002FHan-Jia)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yu Liu](https:\u002F\u002Fliuyu.us\u002F)\u003C\u002Fli> \u003Cli>[Hongsheng Li](https:\u002F\u002Fwww.ee.cuhk.edu.hk\u002F~hsli\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FG-U-N\u002FGen-L-Video?style=social)](https:\u002F\u002Fgithub.com\u002FG-U-N\u002FGen-L-Video) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.18264)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flllyasviel\u002FControlNet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTencentARC\u002FT2I-Adapter), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsail-sg\u002FEditAnything)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fg-u-n.github.io\u002Fprojects\u002Fgen-long-video\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FG-U-N\u002FGen-L-Video\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fglv_colab.ipynb) | 04.06.2023 |\n| First Order Motion Model for Image Animation | Transferring facial movements from video to image | [Aliaksandr Siarohin](https:\u002F\u002Faliaksandrsiarohin.github.io\u002Faliaksandr-siarohin-website\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAliaksandrSiarohin\u002Ffirst-order-model?style=social)](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Ffirst-order-model) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F2019\u002Fhash\u002F31c0b36aef265d9221af80872ceb62f9-Abstract.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Faliaksandrsiarohin.github.io\u002Ffirst-order-model-website\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=u-0cQ-grXBQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAliaksandrSiarohin\u002Ffirst-order-model\u002Fblob\u002Fmaster\u002Fdemo.ipynb) | 04.06.2023 |\n| PolyGen | Approach which models the mesh directly, predicting mesh vertices and faces sequentially using a Transformer-based architecture | \u003Cul>\u003Cli>[Charlie Nash](https:\u002F\u002Fgithub.com\u002Fcharlienash)\u003C\u002Fli> \u003Cli>[Yaroslav Ganin](https:\u002F\u002Fyaroslav.ganin.net\u002F)\u003C\u002Fli> \u003Cli>[Ali Eslami](http:\u002F\u002Farkitus.com\u002F)\u003C\u002Fli> \u003Cli>[Peter Battaglia](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=nQ7Ij30AAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Fdeepmind-research?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Fdeepmind-research\u002Ftree\u002Fmaster\u002Fpolygen) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2002.10880), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1506.03134), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.04887)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fanshulcgm\u002Fpolygen)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FXCrjpIRkVCU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Fdeepmind-research\u002Fblob\u002Fmaster\u002Fpolygen\u002Ftraining.ipynbv) | 02.06.2023 |\n| Parallel WaveGAN | State-of-the-art non-autoregressive models to build your own great vocoder | [Tomoki Hayashi](https:\u002F\u002Fkan-bayashi.github.io\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_137784a250dd.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICASSP40776.2020.9053795) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fkan-bayashi\u002FParallelWaveGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fkan-bayashi\u002FParallelWaveGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.11480), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.06711), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.05106)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fkan-bayashi.github.io\u002FParallelWaveGAN\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Ftacotron2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fespnet\u002Fespnet)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fespnet\u002Fnotebook\u002Fblob\u002Fmaster\u002Fespnet2_tts_realtime_demo.ipynb) | 01.06.2023 |\n| ECON | designed for \"Human digitization from a color image\", which combines the best properties of implicit and explicit representations, to infer high-fidelity 3D clothed humans from in-the-wild images, even with loose clothing or in challenging poses | \u003Cul>\u003Cli>[Yuliang Xiu](https:\u002F\u002Fxiuyuliang.cn\u002F)\u003C\u002Fli> \u003Cli>[Jinlong Yang](https:\u002F\u002Fis.mpg.de\u002F~jyang)\u003C\u002Fli> \u003Cli>[Xu Cao](https:\u002F\u002Fxucao-42.github.io\u002Fhomepage\u002F)\u003C\u002Fli> \u003Cli>[Dimitrios Tzionas](https:\u002F\u002Fps.is.mpg.de\u002F~dtzionas)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ead37f141dbf.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52729.2023.00057) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYuliangXiu\u002FECON?style=social)](https:\u002F\u002Fgithub.com\u002FYuliangXiu\u002FECON) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.07422)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FVqa7KBGRyk)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYuliangXiu\u002FECON\u002Fblob\u002Fmaster\u002Fdocs\u002Finstallation-docker.md)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkwan3854\u002FCEB_ECON), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxucao-42\u002Fbilateral_normal_integration), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FProject-Splinter\u002FMonoPortDataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuangyangyi\u002FTeCH), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhuangyangyi\u002FTeCH), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyfeng95\u002FPIXIE)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F1451sjr\u002Fecon_explicit_clothed_humans_optimized_via_normal\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002Fyuliangxiu)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FsbWZbTf6ZYk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FSDVfCeaI4AY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5PEd_p90kS0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FMDFvV7y5Qgk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1YRgwoRCZIrSB2e7auEWFyG10Xzjbrbno) | 31.05.2023 |\n| MMS | The Massively Multilingual Speech project expands speech technology from about 100 languages to over 1000 by building a single multilingual speech recognition model supporting over 1100 languages, language identification models able to identify over 4000 languages, pretrained models supporting over 1400 languages, and text-to-speech models for over 1100 languages | \u003Cul>\u003Cli>[Vineel Pratap](https:\u002F\u002Fgithub.com\u002Fvineelpratap)\u003C\u002Fli> \u003Cli>[Andros Tjandra](https:\u002F\u002Fgithub.com\u002Fandrostj)\u003C\u002Fli> \u003Cli>[Bowen Shi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=xqyoorYAAAAJ)\u003C\u002Fli> \u003Cli>[Paden Tomasello](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=sBtWMGYAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Arun Babu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=oJfoTakAAAAJ)\u003C\u002Fli> \u003Cli>[Sayani Kundu](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsayani-kundu)\u003C\u002Fli> \u003Cli>[Ali Elkahky](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=KB3S8RoAAAAJ)\u003C\u002Fli> \u003Cli>[Zhaoheng Ni](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=SYFMSNsAAAAJ)\u003C\u002Fli> \u003Cli>[Apoorv Vyas](https:\u002F\u002Fapoorv2904.github.io\u002F)\u003C\u002Fli> \u003Cli>[Maryam Fazel-Zarandi](https:\u002F\u002Fwww.maryamfazel.com\u002F)\u003C\u002Fli> \u003Cli>[Alexei Baevski](https:\u002F\u002Fgithub.com\u002Falexeib)\u003C\u002Fli> \u003Cli>[Yossi Adi](https:\u002F\u002Fwww.cs.huji.ac.il\u002F~adiyoss\u002F)\u003C\u002Fli> \u003Cli>[Xiaohui Zhang](https:\u002F\u002Fgithub.com\u002Fxiaohui-zhang)\u003C\u002Fli> \u003Cli>[Wei-Ning Hsu](https:\u002F\u002Fwnhsu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Alexis Conneau](https:\u002F\u002Fgithub.com\u002Faconneau)\u003C\u002Fli> \u003Cli>[Michael Auli](https:\u002F\u002Fgithub.com\u002Fmichaelauli)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Ffairseq?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffairseq\u002Ftree\u002Fmain\u002Fexamples\u002Fmms) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2305.13516)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmain\u002Fen\u002Fmodel_doc\u002Fmms), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Ffacebook\u002Fmms-cclms\u002F), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fmms_adapters)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.facebook.com\u002Fblog\u002Fmultilingual-model-speech-recognition\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGEzxHxWys2s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fg06agCmxS7I)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Ffairseq\u002Fblob\u002Fmain\u002Fexamples\u002Fmms\u002Fasr\u002Ftutorial\u002FMMS_ASR_Inference_Colab.ipynb) | 26.05.2023 |\n| FAB | Flow AIS Bootstrap uses AIS to generate samples in regions where the flow is a poor approximation of the target, facilitating the discovery of new modes | \u003Cul>\u003Cli>[Laurence Midgley](https:\u002F\u002Flollcat.github.io\u002Flaurence-midgley\u002F)\u003C\u002Fli> \u003Cli>[Vincent Stimper](https:\u002F\u002Fis.mpg.de\u002Fperson\u002Fvstimper)\u003C\u002Fli> \u003Cli>[Gregor N. C. Simm](https:\u002F\u002Fwww.gncs.me\u002F)\u003C\u002Fli> \u003Cli>[Bernhard Schölkopf](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=DZ-fHPgAAAAJ)\u003C\u002Fli> \u003Cli>[José Miguel Hernández-Lobato](https:\u002F\u002Fjmhl.org\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Flollcat\u002Ffab-torch?style=social)](https:\u002F\u002Fgithub.com\u002Flollcat\u002Ffab-torch) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.01893)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flollcat\u002Ffab-jax-old), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fannealed_flow_transport)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FxQQXvOWu9nE)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Flollcat\u002Ffab-torch\u002Fblob\u002Fmaster\u002Fexperiments\u002Fgmm\u002Ffab_gmm.ipynb) | 29.04.2023 |\n| CodeFormer | Transformer-based prediction network to model global composition and context of the low-quality faces for code prediction, enabling the discovery of natural faces that closely approximate the target faces even when the inputs are severely degraded | \u003Cul>\u003Cli>[Shangchen Zhou](https:\u002F\u002Fshangchenzhou.com\u002F)\u003C\u002Fli> \u003Cli>[Kelvin Chan](https:\u002F\u002Fckkelvinchan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chongyi Li](https:\u002F\u002Fli-chongyi.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsczhou\u002FCodeFormer?style=social)](https:\u002F\u002Fgithub.com\u002Fsczhou\u002FCodeFormer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.11253)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsamb-t\u002Funleashing-transformers), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepcam-cn\u002Fyolov5-face), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002Ffacexlib)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002Fc573258c38d0a3919d8c1364053c45df-Abstract-Conference.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fshangchenzhou.com\u002Fprojects\u002FCodeFormer\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fd3VDpkXlueI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPtwWu-FugbA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FORtYP8NW4T0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fxc5lKOKBCcg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1m52PNveE4PBhYrecj34cnpEeiHcC5LTb) | 21.04.2023 |\n| Text2Video-Zero | Text-to-Image Diffusion Models are Zero-Shot Video Generators | \u003Cul>\u003Cli>[Levon Khachatryan](https:\u002F\u002Fgithub.com\u002Flev1khachatryan)\u003C\u002Fli> \u003Cli>[Andranik Movsisyan](https:\u002F\u002Fgithub.com\u002F19and99)\u003C\u002Fli> \u003Cli>[Vahram Tadevosyan](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fvtadevosian)\u003C\u002Fli> \u003Cli>[Roberto Henschel](https:\u002F\u002Fgithub.com\u002Frob-hen)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Zhangyang Wang](https:\u002F\u002Fwww.ece.utexas.edu\u002Fpeople\u002Ffaculty\u002Fatlas-wang)\u003C\u002Fli> \u003Cli>[Shant Navasardyan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=VJSh59sAAAAJ)\u003C\u002Fli> \u003Cli>[Humphrey Shi](https:\u002F\u002Fwww.humphreyshi.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_98f16e41baef.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.01462) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPicsart-AI-Research\u002FText2Video-Zero?style=social)](https:\u002F\u002Fgithub.com\u002FPicsart-AI-Research\u002FText2Video-Zero) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.13439), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.01341), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.17604)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdbolya\u002Ftomesd), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJiauZhang\u002FText2Video-Zero), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcamenduru\u002Ftext2video-zero-colab), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FSHI-Labs\u002FText2Video-Zero-sd-webui)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fdiffusers\u002Fapi\u002Fpipelines\u002Ftext_to_video_zero)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftext2video-zero.github.io\u002F)\u003C\u002Fli>\u003Cli>[video](https:\u002F\u002Fwww.dropbox.com\u002Fs\u002Fuv90mi2z598olsq\u002FText2Video-Zero.MP4)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbeeDJJz-Q0A), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F97-1GYPtz0M)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcamenduru\u002Ftext2video-zero-colab\u002Fblob\u002Fmain\u002Ftext2video_all.ipynb) | 11.04.2023 |\n| Segment Anything | The Segment Anything Model produces high quality object masks from input prompts such as points or boxes, and it can be used to generate masks for all objects in an image | \u003Cul>\u003Cli>[Alexander Kirillov](https:\u002F\u002Falexander-kirillov.github.io\u002F)\u003C\u002Fli> \u003Cli>[Eric Mintun](https:\u002F\u002Fericmintun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nikhila Ravi](https:\u002F\u002Fnikhilaravi.com\u002F)\u003C\u002Fli> \u003Cli>[Hanzi Mao](https:\u002F\u002Fhanzimao.me\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chloé Rolland](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=n-SnMhoAAAAJ)\u003C\u002Fli> \u003Cli>[Laura Gustafson](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=c8IpF9gAAAAJ)\u003C\u002Fli> \u003Cli>[Tete Xiao](https:\u002F\u002Ftetexiao.com\u002F)\u003C\u002Fli> \u003Cli>[Spencer Whitehead](https:\u002F\u002Fwww.spencerwhitehead.com\u002F)\u003C\u002Fli> \u003Cli>[Alex Berg](http:\u002F\u002Facberg.com\u002F)\u003C\u002Fli> \u003Cli>[Wan-Yen Lo](https:\u002F\u002Fgithub.com\u002Fwanyenlo)\u003C\u002Fli> \u003Cli>[Piotr Dollár](https:\u002F\u002Fpdollar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ross Girshick](https:\u002F\u002Fwww.rossgirshick.info\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fsegment-anything?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fsegment-anything) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.02643)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fai.facebook.com\u002Fdatasets\u002Fsegment-anything\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.facebook.com\u002Fresearch\u002Fpublications\u002Fsegment-anything\u002F), [\u003Cimg src=\"images\u002Fmeta.svg\" alt=\"meta\" height=20\u002F>](https:\u002F\u002Fai.facebook.com\u002Fblog\u002Fsegment-anything-foundation-model-image-segmentation\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fsegment-anything.com\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2O_vecl28OA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfVeW9a6wItM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FFjYE0tKWOiY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fsegment-anything\u002Fblob\u002Fmain\u002Fnotebooks\u002Fpredictor_example.ipynb) | 10.04.2023 |\n| FollowYourPose | Two-stage training scheme that can utilize image pose pair and pose-free video datasets and the pre-trained text-to-image model to obtain the pose-controllable character videos | \u003Cul>\u003Cli>[Yue Ma](https:\u002F\u002Fmayuelala.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yingqing He](https:\u002F\u002Fyingqinghe.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002Facademic\u002F)\u003C\u002Fli> \u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Siran Chen](https:\u002F\u002Fgithub.com\u002FSranc3)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli> \u003Cli>[Xiu Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Xrh1OIUAAAAJ)\u003C\u002Fli> \u003Cli>[Qifeng Chen](https:\u002F\u002Fcqf.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_452d0f81a9fa.png)](https:\u002F\u002Fdoi.org\u002F10.1609\u002Faaai.v38i5.28206) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmayuelala\u002FFollowYourPose?style=social)](https:\u002F\u002Fgithub.com\u002Fmayuelala\u002FFollowYourPose) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2304.01186), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10752)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbryandlee\u002FTune-A-Video)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FYueMafighting\u002FFollowYourPose_v1\u002Ftree\u002Fmain), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FCompVis\u002Fstable-diffusion-v1-4)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ffollow-your-pose.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmayuelala)\u003C\u002Fli>\u003Cli>[video](https:\u002F\u002Funderline.io\u002Flecture\u002F91712-follow-your-pose-pose-guided-text-to-video-generation-using-pose-free-videos)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmayuelala\u002FFollowYourPose\u002Fblob\u002Fmain\u002Fquick_demo.ipynb) | 07.04.2023 |\n| EVA3D | High-quality unconditional 3D human generative model that only requires 2D image collections for training | \u003Cul>\u003Cli>[Fangzhou Hong](https:\u002F\u002Fhongfz16.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhaoxi Chen](https:\u002F\u002Ffrozenburning.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yushi Lan](https:\u002F\u002Fgithub.com\u002FNIRVANALAN)\u003C\u002Fli> \u003Cli>[Liang Pan](https:\u002F\u002Fgithub.com\u002Fpaul007pl)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhongfz16\u002FEVA3D?style=social)](https:\u002F\u002Fgithub.com\u002Fhongfz16\u002FEVA3D) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.04888)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fhongfz16.github.io\u002Fprojects\u002FEVA3D.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJNV0FJ0aDWM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FM-kyvzTQrBI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fhongfz16\u002FEVA3D\u002Fblob\u002Fmain\u002Fnotebook\u002FEVA3D_Demo.ipynb) | 06.04.2023 |\n| Stable Dreamfusion | Using a pretrained 2D text-to-image diffusion model to perform text-to-3D synthesis | \u003Cul>\u003Cli>[Jiaxiang Tang](https:\u002F\u002Fme.kiui.moe\u002F)\u003C\u002Fli> \u003Cli>[Ben Poole](https:\u002F\u002Fcs.stanford.edu\u002F~poole\u002F)\u003C\u002Fli> \u003Cli>[Ajay Jain](https:\u002F\u002Fajayj.com\u002F)\u003C\u002Fli> \u003Cli>[Jon Barron](https:\u002F\u002Fjonbarron.info\u002F)\u003C\u002Fli> \u003Cli>[Ben Mildenhall](https:\u002F\u002Fbmild.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fashawkey\u002Fstable-dreamfusion?style=social)](https:\u002F\u002Fgithub.com\u002Fashawkey\u002Fstable-dreamfusion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.14988)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fashawkey\u002Ftorch-ngp), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhoffstadt\u002FDearPyGui)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Frunwayml\u002Fstable-diffusion-v1-5)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fdreamfusion3d.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fpytorch.org\u002Fdocs\u002Fstable\u002Fcpp_extension.html#torch.utils.cpp_extension.load)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuM5NPodZZ1U?t=219), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FzWD5ZR5GtJM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FL3G0dx1Q0R8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdIgDbBTztUM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1MXT3yfOFvO0ooKEfiUUvTKwUkrrlCHpF) | 04.04.2023 |\n| PIFuHD | Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization | \u003Cul>\u003Cli>[Shunsuke Saito](https:\u002F\u002Fshunsukesaito.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tomas Simon](http:\u002F\u002Fwww.cs.cmu.edu\u002F~tsimon\u002F)\u003C\u002Fli> \u003Cli>[Jason Saragih](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ss-IvjMAAAAJ)\u003C\u002Fli> \u003Cli>[Hanbyul Joo](https:\u002F\u002Fjhugestar.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_bb617c34b8dc.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00016) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fpifuhd?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fpifuhd) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.00452)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuEDqCxvF5yc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=8qnwbbDS8xk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F11z58bl3meSzo6kFqkahMa35G5jmh2Wgt) | 26.03.2023 |\n| VideoReTalking | System to edit the faces of a real-world talking head video according to input audio, producing a high-quality and lip-syncing output video even with a different emotion | \u003Cul>\u003Cli>[Kun Cheng](https:\u002F\u002Fgithub.com\u002Fkunncheng)\u003C\u002Fli> \u003Cli>[Xiaodong Cun](https:\u002F\u002Fvinthony.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yong Zhang](https:\u002F\u002Fyzhang2016.github.io\u002F)\u003C\u002Fli> \u003Cli>[Menghan Xia](https:\u002F\u002Fmenghanxia.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Fei Yin](https:\u002F\u002Ffeiiyin.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mingrui Zhu](https:\u002F\u002Fweb.xidian.edu.cn\u002Fmrzhu\u002Fen\u002Findex.html)\u003C\u002Fli> \u003Cli>[Xuan Wang](https:\u002F\u002Fxuanwangvc.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jue Wang](https:\u002F\u002Fjuewang725.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nannan Wang](https:\u002F\u002Fweb.xidian.edu.cn\u002Fnnwang\u002Fen\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ff47e90dab4c.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3550469.3555399) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FOpenTalker\u002Fvideo-retalking?style=social)](https:\u002F\u002Fgithub.com\u002FOpenTalker\u002Fvideo-retalking) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2211.14758)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdonydchen\u002Fganimation_replicate), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FRenYurui\u002FPIRender), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FOpenTalker\u002FStyleHEAT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFeiiYin\u002FSPI)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fxthemadgenius.medium.com\u002Fmaking-videos-talk-right-syncing-lips-with-sound-using-videoretalking-611428084bbc)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fopentalker.github.io\u002Fvideo-retalking\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002F178krha\u002Fvideoretalking\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FpttsTrQ-fko), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2Lkw8AmmRn0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRJ8YK_K4Ne0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fvinthony\u002Fvideo-retalking\u002Fblob\u002Fmain\u002Fquick_demo.ipynb) | 19.03.2023 |\n| Visual ChatGPT | Connects ChatGPT and a series of Visual Foundation Models to enable sending and receiving images during chatting | \u003Cul>\u003Cli>[Chenfei Wu](https:\u002F\u002Fgithub.com\u002Fchenfei-wu)\u003C\u002Fli> \u003Cli>[Shengming Yin](https:\u002F\u002Fgithub.com\u002Fshengming-yin)\u003C\u002Fli> \u003Cli>[Weizhen Qi](https:\u002F\u002Fgithub.com\u002FWeizhenQ)\u003C\u002Fli> \u003Cli>[Xiaodong Wang](https:\u002F\u002Fwang-xiaodong1899.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Zecheng Tang](https:\u002F\u002Fgithub.com\u002FCODINNLG)\u003C\u002Fli> \u003Cli>[Nan Duan](https:\u002F\u002Fnanduan.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002Fvisual-chatgpt?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002Fvisual-chatgpt) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2303.04671)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fhwchase17\u002Flangchain), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flllyasviel\u002FControlNet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftimothybrooks\u002Finstruct-pix2pix), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftimojl\u002Fclipseg)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0UfXlFUwLms), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F7YEiEyfPF5U)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F11BtP3h-w0dZjA-X8JsS9_eo8OeGYvxXB) | 15.03.2023 |\n| Tune-A-Video | One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation | \u003Cul>\u003Cli>[Jay Zhangjie Wu](https:\u002F\u002Fzhangjiewu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yixiao Ge](https:\u002F\u002Fgeyixiao.com\u002F)\u003C\u002Fli> \u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Stan Weixian Lei](https:\u002F\u002Fgithub.com\u002FStanLei52)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yuchao Gu](https:\u002F\u002Fycgu.site\u002F)\u003C\u002Fli> \u003Cli>[Yufei Shi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=rpnlkwEAAAAJ)\u003C\u002Fli> \u003Cli>[Wynne Hsu](https:\u002F\u002Fwww.comp.nus.edu.sg\u002F~whsu\u002F)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli> \u003Cli>[Xiaohu Qie](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=mk-F69UAAAAJ)\u003C\u002Fli> \u003Cli>[Mike Zheng Shou](https:\u002F\u002Fsites.google.com\u002Fview\u002Fshowlab)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_40d4b24cbec4.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV51070.2023.00701) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fshowlab\u002FTune-A-Video?style=social)](https:\u002F\u002Fgithub.com\u002Fshowlab\u002FTune-A-Video) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.11565), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10752)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FTune-A-Video-library), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fstabilityai\u002Fstable-diffusion-2-1), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fsd-dreambooth-library)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftuneavideo.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuzF6CTtjn-g), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuUlp1_ExsGQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fshowlab\u002FTune-A-Video\u002Fblob\u002Fmain\u002Fnotebooks\u002FTune-A-Video.ipynb) | 23.02.2023 |\n| GPEN | GAN Prior Embedded Network for Blind Face Restoration in the Wild | \u003Cul>\u003Cli>[Tao Yang](https:\u002F\u002Fcg.cs.tsinghua.edu.cn\u002Fpeople\u002F~tyang\u002F)\u003C\u002Fli> \u003Cli>[Peiran Ren](https:\u002F\u002Fscholar.google.com\u002Fcitations?&user=x5dEuxsAAAAJ)\u003C\u002Fli> \u003Cli>[Xuansong Xie](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=M0Ei1zkAAAAJ)\u003C\u002Fli> \u003Cli>[Lei Zhang](http:\u002F\u002Fwww4.comp.polyu.edu.hk\u002F~cslzhang\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyangxy\u002FGPEN?style=social)](https:\u002F\u002Fgithub.com\u002Fyangxy\u002FGPEN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.06070)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fvision.aliyun.com\u002Fexperience\u002Fdetail?spm=a211p3.14020179.J_7524944390.17.66cd4850wVDkUQ&tagName=facebody&children=EnhanceFace)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbiubug6\u002FPytorch_Retinaface), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyangxy\u002FGPEN\u002Fblob\u002Fmain\u002FGPEN.ipynb) | 15.02.2023 |\n| PyMAF-X | Кegression-based approach to recovering parametric full-body models from monocular images | \u003Cul>\u003Cli>[Hongwen Zhang](https:\u002F\u002Fhongwenzhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yating Tian](https:\u002F\u002Fgithub.com\u002Ftinatiansjz)\u003C\u002Fli> \u003Cli>[Yuxiang Zhang](https:\u002F\u002Fzhangyux15.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mengcheng Li](https:\u002F\u002Fgithub.com\u002FDw1010)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Liang An](https:\u002F\u002Fanl13.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhenan Sun](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fusers\u002Fznsun\u002F)\u003C\u002Fli> \u003Cli>[Yebin Liu](https:\u002F\u002Fwww.liuyebin.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_17890a89ad9c.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTPAMI.2023.3271691) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FHongwenZhang\u002FPyMAF-X?style=social)](https:\u002F\u002Fgithub.com\u002FHongwenZhang\u002FPyMAF-X) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.06400)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHongwenZhang\u002FDaNet-DensePose2SMPL), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FDensePose), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMicrosoft\u002Fhuman-pose-estimation.pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FMeshGraphormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fleoxiaobin\u002Fdeep-high-resolution-net.pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.liuyebin.com\u002Fpymaf-x\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FylOB0wCeV34)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F13Iytx1Hb0ZryEwbJdpXBW9ggDxs2Y-tL) | 14.02.2023 |\n| Disco Diffusion | A frankensteinian amalgamation of notebooks, models and techniques for the generation of AI Art and Animations | \u003Cul>\u003Cli>[Max Ingham](https:\u002F\u002Fgithub.com\u002Fsomnai-dreams)\u003C\u002Fli> \u003Cli>[Adam Letts](https:\u002F\u002Flinktr.ee\u002Fgandamu)\u003C\u002Fli> \u003Cli>[Daniel Russell](https:\u002F\u002Fgithub.com\u002Frusselldc)\u003C\u002Fli> \u003Cli>[Chigozie Nri](https:\u002F\u002Fgithub.com\u002Fchigozienri)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Falembics\u002Fdisco-diffusion?style=social)](https:\u002F\u002Fgithub.com\u002Falembics\u002Fdisco-diffusion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fguided-diffusion)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_DtWfh9oS54), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FgWxmtdZL8FE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyVJB6oD0_gM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Falembics\u002Fdisco-diffusion\u002Fblob\u002Fmain\u002FDisco_Diffusion.ipynb) | 11.02.2023 |\n| GrooVAE | Some applications of machine learning for generating and manipulating beats and drum performances | \u003Cul>\u003Cli>[Jon Gillick](https:\u002F\u002Fwww.jongillick.com\u002F)\u003C\u002Fli> \u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagenta\u002Fmagenta?style=social)](https:\u002F\u002Fgithub.com\u002Fmagenta\u002Fmagenta\u002Ftree\u002Fmain\u002Fmagenta\u002Fmodels\u002Fmusic_vae) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.06118)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fg.co\u002Fmagenta\u002Fgroovae)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fg.co\u002Fmagenta\u002Fgroove-datasets)\u003C\u002Fli>\u003Cli>[web app](https:\u002F\u002Fgroove-drums.glitch.me\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=x2YLmXzovDo)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fmagenta-demos\u002Fblob\u002Fmaster\u002Fcolab-notebooks\u002FGrooVAE.ipynb) | 02.02.2023 |\n| Multitrack MusicVAE | The models in this notebook are capable of encoding and decoding single measures of up to 8 tracks, optionally conditioned on an underlying chord | \u003Cul>\u003Cli>[Ian Simon](https:\u002F\u002Fgithub.com\u002Fiansimon)\u003C\u002Fli> \u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Colin Raffel](https:\u002F\u002Fcolinraffel.com\u002F\u002F)\u003C\u002Fli> \u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli> \u003Cli>[Douglas Eck](https:\u002F\u002Fgithub.com\u002Fdouglaseck)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1806.00195)\u003C\u002Fli>\u003Cli>[blog post](http:\u002F\u002Fg.co\u002Fmagenta\u002Fmultitrack)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmagenta\u002Fmagenta-demos\u002Fblob\u002Fmaster\u002Fcolab-notebooks\u002FMultitrack_MusicVAE.ipynb) | 02.02.2023 |\n| MusicVAE | A Hierarchical Latent Vector Model for Learning Long-Term Structure in Music | \u003Cul>\u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli> \u003Cli>[Colin Raffel](https:\u002F\u002Fcolinraffel.com\u002F\u002F)\u003C\u002Fli> \u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli> \u003Cli>[Douglas Eck](https:\u002F\u002Fgithub.com\u002Fdouglaseck)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1803.05428)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fg.co\u002Fmagenta\u002Fmusic-vae)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmagenta.tensorflow.org\u002Fmusic-vae)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLBUMAYA6kvGU8Cgqh709o5SUvo-zHGTxr)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmagenta\u002Fmagenta-demos\u002Fblob\u002Fmaster\u002Fcolab-notebooks\u002FMusicVAE.ipynb) | 02.02.2023 |\n| Learning to Paint | Learning to Paint With Model-based Deep Reinforcement Learning | [Manuel Romero](https:\u002F\u002Fmrm8488.github.io\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_6b548051740c.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV.2019.00880) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.04411)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Freinforcementlearning\u002Fcomments\u002Fb5lpfl\u002Flearning_to_paint_with_modelbased_deep\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=YmOgKZ5oipk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmrm8488\u002Fshared_colab_notebooks\u002Fblob\u002Fmaster\u002Fcustom_learningtopaint.ipynb) | 01.02.2023 |\n| LORA | Low-Rank Adaptation, which freezes the pre-trained model weights and injects trainable rank decomposition matrices into each layer of the Transformer architecture, greatly reducing the number of trainable parameters for downstream tasks | \u003Cul>\u003Cli>[Edward Hu](https:\u002F\u002Fedwardjhu.com\u002F)\u003C\u002Fli> \u003Cli>[Yelong Shen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=S6OFEFEAAAAJ)\u003C\u002Fli> \u003Cli>[Phillip Wallis](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=8IqHSXYAAAAJ)\u003C\u002Fli> \u003Cli>[Zeyuan Allen-Zhu](http:\u002F\u002Fzeyuan.allen-zhu.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yuanzhi Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=aHtfItQAAAAJ)\u003C\u002Fli> \u003Cli>[Shean Wang](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fshean-wang-18a20841)\u003C\u002Fli> \u003Cli>[Lu Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=JWJ_SNcAAAAJ)\u003C\u002Fli> \u003Cli>[Weizhu Chen](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fwzchen\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fcloneofsimo\u002Flora?style=social)](https:\u002F\u002Fgithub.com\u002Fcloneofsimo\u002Flora) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.09685), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.12242), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.01618), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.05744), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.04647)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FLoRA)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Flora-library\u002FLoRA-DreamBooth-Training-UI), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Flora)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@shelikohan\u002Flow-rank-adapter-lora-explained-0d3677395639), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Ftowardsdatascience.com\u002Funderstanding-lora-low-rank-adaptation-for-finetuning-large-models-936bce1a07c6\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Floralib\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLocalLLaMA\u002Fcomments\u002F1diw5fb\u002Funderstanding_lora_a_5minute_visual_guide_to\u002F), [\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FLocalLLaMA\u002Fcomments\u002F17z91wk\u002Fpractical_tips_for_finetuning_llms_using_lora\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdA-NhCtrrVE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Ft509sv5MT0w), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FKEv-F5UkhxU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDhRoTONcyZE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPXWYUTMt-AU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FX4VvO3G6_vw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1iSFDpRBKEWr2HLlz243rbym3J2X95kcy) | 30.01.2023 |\n| Instant-NGP | Instant Neural Graphics Primitives with a Multiresolution Hash Encoding | \u003Cul>\u003Cli>[Thomas Müller](https:\u002F\u002Ftom94.net\u002F)\u003C\u002Fli> \u003Cli>[Alex Evans](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Falex-evans)\u003C\u002Fli> \u003Cli>[Christoph Schied](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Fchristoph-schied)\u003C\u002Fli> \u003Cli>[Alexander Keller](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Falex-keller)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9bc55e3acb8c.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3528223.3530127) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNVlabs\u002Finstant-ngp?style=social)](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Finstant-ngp) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.05989)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fdeveloper.nvidia.com\u002Fblog\u002Fgetting-started-with-nvidia-instant-nerfs\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Ftiny-cuda-nn), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FIDLabMedia\u002Flarge-lightfields-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnickponline\u002Fdd-nerf-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Focornut\u002Fimgui), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnothings\u002Fstb)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnvlabs.github.io\u002Finstant-ngp\u002F)\u003C\u002Fli>\u003Cli>[tutorial](https:\u002F\u002Fwww.nvidia.com\u002Fen-us\u002Fon-demand\u002Fsession\u002Fsiggraph2022-sigg22-s-16\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fj8tMk-GE8hY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8GbENSmdVeE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDJ2hcC1orc4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fz3-fjYzd0BA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FNVlabs\u002Finstant-ngp\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finstant_ngp.ipynb) | 18.01.2023 |\n| Fourier Feature Networks | Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains | \u003Cul>\u003Cli>[Matthew Tancik](https:\u002F\u002Fwww.matthewtancik.com\u002F)\u003C\u002Fli> \u003Cli>[Pratul Srinivasan](https:\u002F\u002Fpratulsrinivasan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ben Mildenhall](https:\u002F\u002Fbmild.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sara Fridovich-Keil](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~sfk\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Nithin Raghavan](https:\u002F\u002Fcseweb.ucsd.edu\u002F\u002F~n2raghavan\u002F)\u003C\u002Fli> \u003Cli>[Utkarsh Singhal](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=lvA86MYAAAAJ)\u003C\u002Fli> \u003Cli>[Ravi Ramamoorthi](https:\u002F\u002Fcseweb.ucsd.edu\u002F\u002F~ravir\u002F)\u003C\u002Fli> \u003Cli>[Jon Barron](https:\u002F\u002Fjonbarron.info\u002F)\u003C\u002Fli> \u003Cli>[Ren Ng](https:\u002F\u002Fwww2.eecs.berkeley.edu\u002FFaculty\u002FHomepages\u002Fyirenng.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftancik\u002Ffourier-feature-networks?style=social)](https:\u002F\u002Fgithub.com\u002Ftancik\u002Ffourier-feature-networks) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1806.07572)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2020\u002Fhash\u002F55053683268957697aa39fba6f231c68-Abstract.html), [\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F2007\u002Fhash\u002F013a006f03dbc5392effeb8f18fda755-Abstract.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fbmild.github.io\u002Ffourfeat\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FnVA6K6Sn2S4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftancik\u002Ffourier-feature-networks\u002Fblob\u002Fmaster\u002FDemo.ipynb) | 17.01.2023 |\n| HybrIK | Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation | \u003Cul>\u003Cli>[Jiefeng Li](https:\u002F\u002Fjeffli.site\u002F)\u003C\u002Fli> \u003Cli>[Chao Xu](https:\u002F\u002Fwww.isdas.cn\u002F)\u003C\u002Fli> \u003Cli>[Zhicun Chen](https:\u002F\u002Fgithub.com\u002Fchenzhicun)\u003C\u002Fli> \u003Cli>[Siyuan Bian](https:\u002F\u002Fgithub.com\u002Fbiansy000)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lixin Yang](https:\u002F\u002Flixiny.github.io\u002F)\u003C\u002Fli> \u003Cli>[Cewu Lu](https:\u002F\u002Fwww.mvig.org\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_e4ff3e3e581a.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00339) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FJeff-sjtu\u002FHybrIK?style=social)](https:\u002F\u002Fgithub.com\u002FJeff-sjtu\u002FHybrIK) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.14672)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmks0601\u002F3DMPPE_POSENET_RELEASE)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fjeffli.site\u002FHybrIK\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002F3d-human-pose-estimation-on-3dpw?p=hybrik-a-hybrid-analytical-neural-inverse)\u003C\u002Fli>\u003Cli>[supp](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fsupplemental\u002FLi_HybrIK_A_Hybrid_CVPR_2021_supplemental.zip)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FtvwnXXH7xIw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1n41l7I2NxWseuruVQEU8he2XqzSXhu2f) | 01.01.2023 |\n| Composable-Diffusion | Method can generate scenes at test time that are substantially more complex than those seen in training, composing sentence descriptions, object relations, human facial attributes, and even generalizing to new combinations that are rarely seen in the real world | \u003Cul>\u003Cli>[Nan Liu](https:\u002F\u002Fnanliu.io)\u003C\u002Fli> \u003Cli>[Shuang Li](https:\u002F\u002Fshuangli01.github.io)\u003C\u002Fli> \u003Cli>[Yilun Du](https:\u002F\u002Fyilundu.github.io)\u003C\u002Fli> \u003Cli>[Antonio Torralba](https:\u002F\u002Fwww.csail.mit.edu\u002Fperson\u002Fantonio-torralba)\u003C\u002Fli> \u003Cli>[Joshua Tenenbaum](http:\u002F\u002Fweb.mit.edu\u002Fcocosci\u002Fjosh.html)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c36afe809fc9.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-19790-1_26) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fenergy-based-model\u002FCompositional-Visual-Generation-with-Composable-Diffusion-Models-PyTorch?style=social)](https:\u002F\u002Fgithub.com\u002Fenergy-based-model\u002FCompositional-Visual-Generation-with-Composable-Diffusion-Models-PyTorch) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.01714)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fpoint-e), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fimproved-diffusion)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FShuang59\u002FComposable-Diffusion)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fenergy-based-model.github.io\u002FCompositional-Visual-Generation-with-Composable-Diffusion-Models\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002Fxwplfv\u002Fand_prompt_combinations_just_landed_in\u002F), [\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002Fxf5jow\u002Fcompositional_diffusion\u002F), [\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002Fxoq7ik\u002Fcomposable_diffusion_a_new_development_to_greatly\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FmQzF6BDKes4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fenergy-based-model\u002FCompositional-Visual-Generation-with-Composable-Diffusion-Models-PyTorch\u002Fblob\u002Fmain\u002Fnotebooks\u002Fdemo.ipynb) | 23.12.2022 |\n| Score Jacobian Chaining | Apply chain rule on the learned gradients, and back-propagate the score of a diffusion model through the Jacobian of a differentiable renderer, which we instantiate to be a voxel radiance field | \u003Cul>\u003Cli>[Haochen Wang](https:\u002F\u002Fwhc.is\u002F)\u003C\u002Fli> \u003Cli>[Xiaodan Du](https:\u002F\u002Fxiaodan.io\u002F)\u003C\u002Fli> \u003Cli>[Jiahao Li](https:\u002F\u002Fjiahao.ai\u002F)\u003C\u002Fli> \u003Cli>[Raymond Yeh](https:\u002F\u002Fraymond-yeh.com\u002F)\u003C\u002Fli> \u003Cli>[Greg Shakhnarovich](https:\u002F\u002Fhome.ttic.edu\u002F~gregory\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a41f02ed2b74.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52729.2023.01214) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fpals-ttic\u002Fsjc?style=social)](https:\u002F\u002Fgithub.com\u002Fpals-ttic\u002Fsjc) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2212.00774), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.00364)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FMirageML\u002Fsjc)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpals.ttic.edu\u002Fp\u002Fscore-jacobian-chaining)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FStableDiffusion\u002Fcomments\u002Fzac8z4\u002Fscore_jacobian_chaining_lifting_pretrained_2d\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FMmDSLc6CjoI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F1oeruRLKoiU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1zixo66UYGl70VOPy053o7IV_YkQt5lCZ) | 05.12.2022 |\n| Demucs | Hybrid Spectrogram and Waveform Source Separation | [Alexandre Défossez](https:\u002F\u002Fai.honu.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fdemucs?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdemucs) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.03600), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.01733), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.05418), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1805.02410)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fadefossez\u002Fmdx21_demucs), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCarlGao4\u002FDemucs-Gui), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkuielab\u002Fmdx-net-submission), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ff90\u002FWave-U-Net)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1dC9nVxk3V_VPjUADsnFu8EiT-xnU1tGH) | 21.11.2022 |\n| FSGAN | Face Swapping GAN for face swapping and reenactment | \u003Cul>\u003Cli>[Yuval Nirkin](https:\u002F\u002Fnirkin.com\u002F)\u003C\u002Fli> \u003Cli>[Yosi Keller](https:\u002F\u002Fyosikeller.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tal Hassner](https:\u002F\u002Ftalhassner.github.io\u002Fhome\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_31e02e091914.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV.2019.00728) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYuvalNirkin\u002Ffsgan?style=social)](https:\u002F\u002Fgithub.com\u002FYuvalNirkin\u002Ffsgan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.05932)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fondyari\u002FFaceForensics)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fodsc.medium.com\u002Ffsgan-subject-agnostic-face-swapping-and-reenactment-2f033b0ea83c)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnirkin.com\u002Ffsgan\u002F), [project](https:\u002F\u002Fnirkin.com\u002Ffsgan-v2\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBsITEVX6hkE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fduo-tHbSdMk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FcfEqjXkCcCI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F4rNUkmEqgng)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FYuvalNirkin\u002Ffsgan\u002Fblob\u002Fmaster\u002Ffsgan\u002Finference\u002Fface_swapping.ipynb) | 16.11.2022 |\n| StyleCLIP | Text-Driven Manipulation of StyleGAN Imager | \u003Cul>\u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zongze Wu](https:\u002F\u002Fgithub.com\u002Fbetterze)\u003C\u002Fli> \u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli> \u003Cli>[Dani Lischinski](https:\u002F\u002Fpages.cs.huji.ac.il\u002Fdanix\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9cadabca3ad7.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00209) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Forpatashnik\u002FStyleCLIP?style=social)](https:\u002F\u002Fgithub.com\u002Forpatashnik\u002FStyleCLIP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.17249), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.12799)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5icI0NgALnQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FPhR1gpXDu0w), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fd1OET63Ulwc), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRAXrwPskNso)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Forpatashnik\u002FStyleCLIP\u002Fblob\u002Fmain\u002Fnotebooks\u002FStyleCLIP_global_torch.ipynb) | 30.10.2022 |\n| AST | Audio Spectrogram Transformer, the first convolution-free, purely attention-based model for audio classification | \u003Cul>\u003Cli>[Yuan Gong](https:\u002F\u002Fyuangongnd.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yu-An Chung](https:\u002F\u002Fiamyuanchung.github.io\u002F)\u003C\u002Fli> \u003Cli>[James Glass](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=pfGI-KcAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_bb66827c48fd.png)](https:\u002F\u002Fdoi.org\u002F10.21437\u002FInterspeech.2021-698) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYuanGongND\u002Fast?style=social)](https:\u002F\u002Fgithub.com\u002FYuanGongND\u002Fast) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.01778), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.06760), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2110.09784), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.01243)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYuanGongND\u002Fltu), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYuanGongND\u002Fssast), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FYuanGongND\u002Fpsla)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Freview-ast-audio-spectrogram-transformer-a108a5775d2f)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FiKqmvNSGuyw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F9vGeIyeRRNQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002FCSRDbqGY0Vw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FYuanGongND\u002Fast\u002Fblob\u002Fmaster\u002Fcolab\u002FAST_Inference_Demo.ipynb) | 18.10.2022 |\n| MotionDiffuse | The first diffusion model-based text-driven motion generation framework, which demonstrates several desired properties over existing methods | \u003Cul>\u003Cli>[Mingyuan Zhang](https:\u002F\u002Fmingyuan-zhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhongang Cai](https:\u002F\u002Fcaizhongang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liang Pan](https:\u002F\u002Fgithub.com\u002Fpaul007pl)\u003C\u002Fli> \u003Cli>[Fangzhou Hong](https:\u002F\u002Fhongfz16.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xinying Guo](https:\u002F\u002Fgxyes.github.io\u002F)\u003C\u002Fli> \u003Cli>[Lei Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jZH2IPYAAAAJ)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmingyuan-zhang\u002FMotionDiffuse?style=social)](https:\u002F\u002Fgithub.com\u002Fmingyuan-zhang\u002FMotionDiffuse) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.15001)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fmingyuan\u002FMotionDiffuse)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fmingyuan-zhang.github.io\u002Fprojects\u002FMotionDiffuse.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FU5PTnw490SA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Dp6VsZp2ozKuu9ccMmsDjyij_vXfCYb3) | 13.10.2022 |\n| VToonify | Leverages the mid- and high-resolution layers of StyleGAN to render high-quality artistic portraits based on the multi-scale content features extracted by an encoder to better preserve the frame details | \u003Cul>\u003Cli>[Shuai Yang](https:\u002F\u002Fwilliamyang1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liming Jiang](https:\u002F\u002Fliming-jiang.com\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cba6c3f73678.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3550454.3555437) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwilliamyang1991\u002FVToonify?style=social)](https:\u002F\u002Fgithub.com\u002Fwilliamyang1991\u002FVToonify) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.11224), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2001.02890)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzllrunning\u002Fface-parsing.PyTorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhujiapeng\u002FLowRankGAN)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FPKUWilliamYang\u002FVToonify), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002FPKUWilliamYang\u002FVToonify\u002Ftree\u002Fmain\u002Fmodels)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fproject\u002Fvtoonify\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0_OmVhDgYuY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](http:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fwilliamyang1991\u002FVToonify\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 07.10.2022 |\n| PyMAF | Pyramidal Mesh Alignment Feedback loop in regression network for well-aligned body mesh recovery and extend it for the recovery of expressive full-body models | \u003Cul>\u003Cli>[Hongwen Zhang](https:\u002F\u002Fhongwenzhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yating Tian](https:\u002F\u002Fgithub.com\u002Ftinatiansjz)\u003C\u002Fli> \u003Cli>[Yuxiang Zhang](https:\u002F\u002Fzhangyux15.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mengcheng Li](https:\u002F\u002Fgithub.com\u002FDw1010)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Liang An](https:\u002F\u002Fanl13.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhenan Sun](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fusers\u002Fznsun\u002F)\u003C\u002Fli> \u003Cli>[Yebin Liu](https:\u002F\u002Fwww.liuyebin.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FHongwenZhang\u002FPyMAF?style=social)](https:\u002F\u002Fgithub.com\u002FHongwenZhang\u002FPyMAF) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.06400), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.16507)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Feft), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHongwenZhang\u002FDaNet-DensePose2SMPL), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FDensePose), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMicrosoft\u002Fhuman-pose-estimation.pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.liuyebin.com\u002Fpymaf-x\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyqEmznSKjYI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FylOB0wCeV34)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F11RXLsH9BdoSCwY6G-IX7KgqDxVoImu6K) | 06.10.2022 |\n| AlphaTensor | Discovering faster matrix multiplication algorithms with reinforcement learning | \u003Cul>\u003Cli>[Alhussein Fawzi](http:\u002F\u002Fwww.alhusseinfawzi.info\u002F)\u003C\u002Fli> \u003Cli>[Matej Balog](http:\u002F\u002Fmatejbalog.eu\u002F)\u003C\u002Fli> \u003Cli>[Aja Huang](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FAja_Huang)\u003C\u002Fli> \u003Cli>[Thomas Hubert](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=WXG0QfMAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Bernardino Romera-Paredes](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fromeraparedes\u002F)\u003C\u002Fli> \u003Cli>[Mohammadamin Barekatain](http:\u002F\u002Fbarekatain.me\u002F)\u003C\u002Fli> \u003Cli>[Alexander Novikov](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jMUkLqwAAAAJ)\u003C\u002Fli> \u003Cli>[Francisco Ruiz](https:\u002F\u002Ffranrruiz.github.io\u002F)\u003C\u002Fli> \u003Cli>[Julian Schrittwieser](https:\u002F\u002Fwww.furidamu.org\u002F)\u003C\u002Fli> \u003Cli>[Grzegorz Swirszcz](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fgrzegorzswirszcz\u002Fhome)\u003C\u002Fli> \u003Cli>[David Silver](https:\u002F\u002Fwww.davidsilver.uk\u002F)\u003C\u002Fli> \u003Cli>[Demis Hassabis](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FDemis_Hassabis)\u003C\u002Fli> \u003Cli>[Pushmeet Kohli](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fpushmeet\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_37c21b7bb6c8.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41586-022-05172-4) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-deepmind\u002Falphatensor?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-deepmind\u002Falphatensor) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fwww.deepmind.com\u002Fblog\u002Fdiscovering-novel-algorithms-with-alphatensor)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3N3Bl5AA5QU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FgpYnDls4PdQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIYgZS2EvnLI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8ILk4Wjo5rc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Falphatensor\u002Fblob\u002Fmaster\u002Fnonequivalence\u002Finspect_factorizations_notebook.ipynb) | 04.10.2022 |\n| Swin2SR | Novel Swin Transformer V2, to improve SwinIR for image super-resolution, and in particular, the compressed input scenario | \u003Cul>\u003Cli>[Marcos Conde](https:\u002F\u002Fmv-lab.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ui-Jin Choi](https:\u002F\u002Fgithub.com\u002FChoiuijin1125)\u003C\u002Fli> \u003Cli>[Maxime Burchi](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=7S_l2eAAAAAJ)\u003C\u002Fli> \u003Cli>[Radu Timofte](https:\u002F\u002Fwww.informatik.uni-wuerzburg.de\u002Fcomputervision\u002Fhome\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_18e61c523853.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-25063-7_42) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmv-lab\u002Fswin2sr?style=social)](https:\u002F\u002Fgithub.com\u002Fmv-lab\u002Fswin2sr) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2209.11345), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.10257), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.11184), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.09883)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcszn\u002FKAIR\u002F), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmv-lab\u002FAISP), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FSwin-Transformer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fjjourney1125\u002Fswin2sr)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fcode\u002Fjesucristo\u002Fsuper-resolution-demo-swin2sr-official\u002F), [\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fdatasets\u002Fjesucristo\u002Fsuper-resolution-benchmarks), [\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fjinssaa\u002Fofficial-swin2sr-demo-results\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1paPrt62ydwLv2U2eZqfcFsePI4X4WRR1) | 03.10.2022 |\n| Functa | From data to functa: Your data point is a function and you can treat it like one | \u003Cul>\u003Cli>[Emilien Dupont](https:\u002F\u002Femiliendupont.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hyunjik Kim](https:\u002F\u002Fhyunjik11.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ali Eslami](http:\u002F\u002Farkitus.com\u002F)\u003C\u002Fli> \u003Cli>[Danilo Rezende](https:\u002F\u002Fdanilorezende.com\u002Fabout\u002F)\u003C\u002Fli> \u003Cli>[Dan Rosenbaum](https:\u002F\u002Fdanrsm.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepmind\u002Ffuncta?style=social)](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Ffuncta) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.12204)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsxyu\u002Fpixel-nerf), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fjaxline)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fdatasets\u002Fcatalog\u002Fceleb_a_hq)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Ffuncta\u002Fblob\u002Fmain\u002Fmodulation_visualization_colab.ipynb) | 24.09.2022 |\n| DeOldify (photo) | Colorize your own photos! | \u003Cul>\u003Cli>[Jason Antic](https:\u002F\u002Fgithub.com\u002Fjantic)\u003C\u002Fli> \u003Cli>[Matt Robinson](https:\u002F\u002Fgithub.com\u002Fmc-robinson)\u003C\u002Fli> \u003Cli>[María Benavente](https:\u002F\u002Fgithub.com\u002Fmariabg)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjantic\u002FDeOldify?style=social)](https:\u002F\u002Fgithub.com\u002Fjantic\u002FDeOldify) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1805.08318), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.08500)\u003C\u002Fli>\u003Cli>[model](https:\u002F\u002Fdata.deepai.org\u002Fdeoldify\u002FColorizeArtistic_gen.pth)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FTheWayWeWere\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002FDeOldify)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fdeoldify.ai\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fjantic\u002FDeOldify\u002Fblob\u002Fmaster\u002FImageColorizerColab.ipynb) | 19.09.2022 |\n| DeOldify (video) | Colorize your own videos! | [Jason Antic](https:\u002F\u002Fgithub.com\u002Fjantic) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjantic\u002FDeOldify?style=social)](https:\u002F\u002Fgithub.com\u002Fjantic\u002FDeOldify) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1805.08318), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.08500)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Felement-ai-research-lab\u002Fstabilizing-neural-style-transfer-for-video-62675e203e42)\u003C\u002Fli>\u003Cli>[model](https:\u002F\u002Fdata.deepai.org\u002Fdeoldify\u002FColorizeVideo_gen.pth)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FNickelodeons\u002F), [\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fsilentmoviegifs\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002FDeOldify)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fdeoldify.ai\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](http:\u002F\u002Fwww.youtube.com\u002Fwatch?v=l3UXXid04Ys), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](http:\u002F\u002Fwww.youtube.com\u002Fwatch?v=EXn-n2iqEjI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fjantic\u002FDeOldify\u002Fblob\u002Fmaster\u002FVideoColorizerColab.ipynb) | 19.09.2022 |\n| Real-ESRGAN | Extend the powerful ESRGAN to a practical restoration application, which is trained with pure synthetic data | \u003Cul>\u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liangbin Xie](https:\u002F\u002Fliangbinxie.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chao Dong](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=OSDCB0UAAAAJ)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c3c44214a50a.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCVW54120.2021.00217) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fxinntao\u002FReal-ESRGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fxinntao\u002FReal-ESRGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.10833)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002FESRGAN), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002Ffacexlib), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002FHandyView), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTencent\u002Fncnn), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnihui\u002Fwaifu2x-ncnn-vulkan)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1k2Zod6kSHEvraybHl50Lys0LerhyTMCo) | 18.09.2022 |\n| IDE-3D | Interactive Disentangled Editing for High-Resolution 3D-aware Portrait Synthesis | \u003Cul>\u003Cli>[Jingxiang Sun](https:\u002F\u002Fmrtornado24.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xuan Wang](https:\u002F\u002Fxuanwangvc.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yichun Shi](https:\u002F\u002Fseasonsh.github.io\u002F)\u003C\u002Fli> \u003Cli>[Lizhen Wang](https:\u002F\u002Flizhenwangt.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jue Wang](https:\u002F\u002Fjuewang725.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yebin Liu](http:\u002F\u002Fwww.liuyebin.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_952a6250d74e.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3550454.3555506) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FMrTornado24\u002FIDE-3D?style=social)](https:\u002F\u002Fgithub.com\u002FMrTornado24\u002FIDE-3D) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.15517), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Feg3d), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan3)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FKj5XY_J2Alk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FMrTornado24\u002FIDE-3D\u002Fblob\u002Fmain\u002Finversion\u002Fnotebooks\u002Finference_playground.ipynb) | 08.09.2022 |\n| Decision Transformers | An architecture that casts the problem of RL as conditional sequence modeling | \u003Cul>\u003Cli>[Lili Chen](http:\u002F\u002Fwww.lilichen.me\u002F)\u003C\u002Fli> \u003Cli>[Kevin Lu](https:\u002F\u002Fkzl.github.io\u002F)\u003C\u002Fli> \u003Cli>[Aravind Rajeswaran](https:\u002F\u002Faravindr93.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kimin Lee](https:\u002F\u002Fsites.google.com\u002Fview\u002Fkiminlee)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Aditya Grover](https:\u002F\u002Faditya-grover.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michael Laskin](https:\u002F\u002Fwww.mishalaskin.com\u002F)\u003C\u002Fli> \u003Cli>[Pieter Abbeel](http:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~pabbeel\u002F)\u003C\u002Fli> \u003Cli>[Aravind Srinivas](https:\u002F\u002Fgithub.com\u002Faravindsrinivas)\u003C\u002Fli> \u003Cli>[Igor Mordatch](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Vzr1RukAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fkzl\u002Fdecision-transformer?style=social)](https:\u002F\u002Fgithub.com\u002Fkzl\u002Fdecision-transformer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.01345)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fmodels?other=gym-continous-control), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fedbeeching\u002Fdecision-transformer-gym-hopper-expert), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Fdecision_transformer)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsites.google.com\u002Fberkeley.edu\u002Fdecision-transformer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FAutoregressive_model)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fk08N5a0gG0A), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-buULmf7dec), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F83QN9S-0I84), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fw4Bw8WYL8Ps)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1K3UuajwoPY1MzRKNkONNRS3gS5DxZ-qF) | 06.09.2022 |\n| textual-inversion | An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion | \u003Cul>\u003Cli>[Rinon Gal](https:\u002F\u002Frinongal.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuval Atzmon](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Fyuval-atzmon)\u003C\u002Fli> \u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Amit Bermano](https:\u002F\u002Fwww.cs.tau.ac.il\u002F~amberman\u002F)\u003C\u002Fli> \u003Cli>[Gal Chechik](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Fgal-chechik)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Frinongal\u002Ftextual_inversion?style=social)](https:\u002F\u002Fgithub.com\u002Frinongal\u002Ftextual_inversion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.01618)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftextual-inversion.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Ff3oXa7_SYek), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FopD_H9bED9Y)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Frinongal\u002Ftextual_inversion\u002Fblob\u002Fmaster\u002Fscripts\u002Flatent_imagenet_diffusion.ipynb) | 21.08.2022 |\n| StyleGAN-Human | A Data-Centric Odyssey of Human Generation | \u003Cul>\u003Cli>[Jianglin Fu](https:\u002F\u002Fgithub.com\u002FarleneF)\u003C\u002Fli> \u003Cli>[Shikai Li](https:\u002F\u002Fgithub.com\u002Fleeskyed)\u003C\u002Fli> \u003Cli>[Yuming Jiang](https:\u002F\u002Fyumingj.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kwan-Yee Lin](https:\u002F\u002Fkwanyeelin.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chen Qian](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=AerkT0YAAAAJ)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli> \u003Cli>[Wayne Wu](https:\u002F\u002Fwywu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_19bf517b7e29.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-19787-1_1) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fstylegan-human\u002Fstylegan-human?style=social)](https:\u002F\u002Fgithub.com\u002Fstylegan-human\u002Fstylegan-human) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.11823)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan3)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fstylegan-human.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fdataset\u002Fmarket-1501)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FnIrb9hwsdcI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F86b49sCz0Gg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fg3nmM6MdxwY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fp2uwqh_SFL8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1sgxoDM55iM07FS54vz9ALg1XckiYA2On) | 19.08.2022 |\n| Make-A-Scene | Scene-Based Text-to-Image Generation with Human Priors | \u003Cul>\u003Cli>[Oran Gafni](https:\u002F\u002Fgithub.com\u002Fogafni)\u003C\u002Fli> \u003Cli>[Adam Polyak](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=CP62OTMAAAAJ)\u003C\u002Fli> \u003Cli>[Oron Ashual](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=CUA9JCkAAAAJ)\u003C\u002Fli> \u003Cli>[Shelly Sheynin](https:\u002F\u002Fgithub.com\u002Fshellysheynin)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Devi Parikh](https:\u002F\u002Ffaculty.cc.gatech.edu\u002F~parikh\u002F)\u003C\u002Fli> \u003Cli>[Yaniv Taigman](https:\u002F\u002Fytaigman.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCasualGANPapers\u002FMake-A-Scene?style=social)](https:\u002F\u002Fgithub.com\u002FCasualGANPapers\u002FMake-A-Scene) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.13131)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FZM06MjPdoxw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1SPyQ-epTsAOAu8BEohUokN4-b5RM_TnE) | 12.08.2022 |\n| StyleGAN-NADA | Zero-Shot non-adversarial domain adaptation of pre-trained generators | \u003Cul>\u003Cli>[Rinon Gal](https:\u002F\u002Frinongal.github.io\u002F)\u003C\u002Fli> \u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haggai Maron](https:\u002F\u002Fhaggaim.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gal Chechik](https:\u002F\u002Fresearch.nvidia.com\u002Fperson\u002Fgal-chechik)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0cd2480f3d79.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3528223.3530164) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Frinongal\u002FStyleGAN-nada?style=social)](https:\u002F\u002Fgithub.com\u002Frinongal\u002FStyleGAN-nada) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.00946), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.17249), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.02699)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch\u002F), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fstylegan-nada.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Frinongal\u002Fstylegan-nada\u002Fblob\u002Fmain\u002Fstylegan_nada.ipynb) | 09.08.2022 |\n| YOLOv7 | Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors | \u003Cul>\u003Cli>[Chien-Yao Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=DkQh4M4AAAAJ)\u003C\u002Fli> \u003Cli>[Alexey Bochkovskiy](http:\u002F\u002Fwww.alexeyab.com\u002F)\u003C\u002Fli> \u003Cli>[Mark Liao](https:\u002F\u002Fwww.iis.sinica.edu.tw\u002Fpages\u002Fliao\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FWongKinYiu\u002Fyolov7?style=social)](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002Fyolov7) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2207.02696)\u003C\u002Fli>\u003Cli>[data](http:\u002F\u002Fimages.cocodataset.org\u002Fannotations\u002Fannotations_trainval2017.zip), [data](http:\u002F\u002Fimages.cocodataset.org\u002Fzips\u002Ftrain2017.zip), [data](http:\u002F\u002Fimages.cocodataset.org\u002Fzips\u002Fval2017.zip), [data](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002Fyolov7\u002Freleases\u002Fdownload\u002Fv0.1\u002Fcoco2017labels-segments.zip)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002Fyolor), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002FPyTorch_YOLOv4), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FWongKinYiu\u002FScaledYOLOv4), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMegvii-BaseDetection\u002FYOLOX), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDingXiaoH\u002FRepVGG), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJUGGHM\u002FOREPA_CVPR2022), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTexasInstruments\u002Fedgeai-yolov5\u002Ftree\u002Fyolo-pose)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Freal-time-object-detection-on-coco?p=yolov7-trainable-bag-of-freebies-sets-new)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PL_Nji0JOuXg2QMohGK7wfzgJ-MavzXRHW), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-QWxJ0j9EY8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FWongKinYiu\u002Fyolov7\u002Fblob\u002Fmain\u002Ftools\u002Fcompare_YOLOv7_vs_YOLOv5m6_half.ipynb) | 09.08.2022 |\n| GLIP | Grounded language-image pre-training model for learning object-level, language-aware, and semantic-rich visual representations | \u003Cul>\u003Cli>[Liunian Harold Li](https:\u002F\u002Fliunian-harold-li.github.io\u002F)\u003C\u002Fli> \u003Cli>[Pengchuan Zhang](https:\u002F\u002Fpzzhang.github.io\u002Fpzzhang\u002F)\u003C\u002Fli> \u003Cli>[Haotian Zhang](https:\u002F\u002Fhaotian-zhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jianwei Yang](https:\u002F\u002Fjwyang.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chunyuan Li](https:\u002F\u002Fchunyuan.li\u002F)\u003C\u002Fli> \u003Cli>[Yiwu Zhong](https:\u002F\u002Fpages.cs.wisc.edu\u002F~yiwuzhong\u002F)\u003C\u002Fli> \u003Cli>[Lijuan Wang](https:\u002F\u002Fgithub.com\u002FLijuanWang)\u003C\u002Fli> \u003Cli>[Lu Yuan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=k9TsUVsAAAAJ)\u003C\u002Fli> \u003Cli>[Lei Zhang](https:\u002F\u002Fwww.leizhang.org\u002F)\u003C\u002Fli> \u003Cli>[Jenq-Neng Hwang](https:\u002F\u002Fpeople.ece.uw.edu\u002Fhwang\u002F)\u003C\u002Fli> \u003Cli>[Kai-Wei Chang](http:\u002F\u002Fweb.cs.ucla.edu\u002F~kwchang\u002F)\u003C\u002Fli> \u003Cli>[Jianfeng Gao](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fjfgao\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_085ff51929b8.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01069) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002FGLIP?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FGLIP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.03857), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.05836), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.01066), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.08790)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fproject\u002Fproject-florence-vl\u002Farticles\u002Fobject-detection-in-the-wild-via-grounded-language-image-pre-training\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgligen\u002FGLIGEN)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fharold\u002FGLIP)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Fglip-grounded-language-image-pre-training-2be2483295b3), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Ftowardsdatascience.com\u002Fglip-introducing-language-image-pre-training-to-object-detection-5ddb601873aa)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fzu1BGQBI4dU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F12x7v-_miN7-SRiziK3Cx4ffJzstBJNqb) | 30.07.2022 |\n| Anycost GAN | Interactive natural image editing | \u003Cul>\u003Cli>[Ji Lin](http:\u002F\u002Flinji.me\u002F)\u003C\u002Fli> \u003Cli>[Richard Zhang](https:\u002F\u002Frichzhang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Frieder Ganz](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=u9ySZkUAAAAJ)\u003C\u002Fli> \u003Cli>[Song Han](https:\u002F\u002Fsonghan.mit.edu\u002F)\u003C\u002Fli> \u003Cli>[Jun-Yan Zhu](https:\u002F\u002Fwww.cs.cmu.edu\u002F~junyanz\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a5493ea31851.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01474) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmit-han-lab\u002Fanycost-gan?style=social)](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fanycost-gan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.03243)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fswitchablenorms\u002FCelebAMask-HQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffyu\u002Flsun)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fhanlab.mit.edu\u002Fprojects\u002Fanycost-gan\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=_yEziPl9AkM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmit-han-lab\u002Fanycost-gan\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fintro_colab.ipynb) | 20.07.2022 |\n| GFPGAN | Towards Real-World Blind Face Restoration with Generative Facial Prior | \u003Cul>\u003Cli>[Xintao Wang](https:\u002F\u002Fxinntao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yu Li](https:\u002F\u002Fyu-li.github.io\u002F)\u003C\u002Fli> \u003Cli>[Honglun Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=KjQLROoAAAAJ)\u003C\u002Fli> \u003Cli>[Ying Shan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=4oXBp9UAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_b133ac9e676e.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00905) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTencentARC\u002FGFPGAN?style=social)](https:\u002F\u002Fgithub.com\u002FTencentARC\u002FGFPGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2101.04061)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002Ffacexlib), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fxinntao\u002FHandyView), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fxinntao.github.io\u002Fprojects\u002Fgfpgan)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1sVsoBd9AjckIXThgtZhGrHRfFI6UUYOo) | 13.07.2022 |\n| EPro-PnP | Generalized End-to-End Probabilistic Perspective-n-Points for Monocular Object Pose Estimation | \u003Cul>\u003Cli>[Hansheng Chen](https:\u002F\u002Flakonik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Pichao Wang](https:\u002F\u002Fwangpichao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Fan Wang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=WCRGTHsAAAAJ)\u003C\u002Fli> \u003Cli>[Wei Tian](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=aYKQn88AAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lu Xiong](https:\u002F\u002Fieeexplore.ieee.org\u002Fauthor\u002F37401835800)\u003C\u002Fli> \u003Cli>[Hao Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=pHN-QIwAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fafa450d6826.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTPAMI.2024.3354997) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftjiiv-cprg\u002FEPro-PnP?style=social)](https:\u002F\u002Fgithub.com\u002Ftjiiv-cprg\u002FEPro-PnP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.13254)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmegvii-research\u002Fpetr), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHuangJunJie2017\u002FBEVDet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffudan-zvg\u002FPolarFormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fzhiqi-li\u002FBEVFormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopen-mmlab\u002Fmmdetection3d)\u003C\u002Fli>\u003Cli>[nuScenes](https:\u002F\u002Fwww.nuscenes.org\u002Fobject-detection?externalData=no&mapData=no&modalities=Camera)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTonBodQ6EUU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftjiiv-cprg\u002FEPro-PnP\u002Fblob\u002Fmain\u002Fdemo\u002Ffit_identity.ipynb) | 12.07.2022 |\n| Text2Human | Text-driven controllable framework for a high-quality and diverse human generation | \u003Cul>\u003Cli>[Yuming Jiang](https:\u002F\u002Fyumingj.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shuai Yang](https:\u002F\u002Fwilliamyang1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haonan Qiu](http:\u002F\u002Fhaonanqiu.com\u002F)\u003C\u002Fli> \u003Cli>[Wayne Wu](https:\u002F\u002Fwywu.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_898be9756148.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3528223.3530104) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyumingj\u002FText2Human?style=social)](https:\u002F\u002Fgithub.com\u002Fyumingj\u002FText2Human) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.15996)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyumingj\u002FDeepFashion-MultiModal), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsamb-t\u002Funleashing-transformers)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fhysts\u002FText2Human), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FCVPR\u002Fdrawings-to-human)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyumingj.github.io\u002Fprojects\u002FText2Human.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FyKh4VORA_E0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRV-g5BlH3Zg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1AVwbqLwMp_Gz3KTCgBTtnGVtXIlCZDPk) | 04.07.2022 |\n| VQ-Diffusion | Based on a VQ-VAE whose latent space is modeled by a conditional variant of the recently developed Denoising Diffusion Probabilistic Model | \u003Cul>\u003Cli>[Shuyang Gu](https:\u002F\u002Fgithub.com\u002Fcientgu)\u003C\u002Fli> \u003Cli>[Dong Chen](http:\u002F\u002Fwww.dongchen.pro\u002F)\u003C\u002Fli> \u003Cli>[Jianmin Bao](https:\u002F\u002Fjianminbao.github.io\u002F)\u003C\u002Fli> \u003Cli>[Fang Wen](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Ffangwen\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Bo Zhang](https:\u002F\u002Fbo-zhang.me\u002F)\u003C\u002Fli> \u003Cli>[Dongdong Chen](http:\u002F\u002Fwww.dongdongchen.bid\u002F)\u003C\u002Fli> \u003Cli>[Lu Yuan](https:\u002F\u002Fscholar.google.com\u002Fcitations?&user=k9TsUVsAAAAJ)\u003C\u002Fli> \u003Cli>[Baining Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=h4kYmRYAAAAJ)\u003C\u002Fli> \u003Cli>[Shuyang Gu](https:\u002F\u002Fgithub.com\u002Fcientgu)\u003C\u002Fli> \u003Cli>[Zhicong Tang](https:\u002F\u002Fgithub.com\u002Fzzctan)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_3b7ddc97ebfd.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01043) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002FVQ-Diffusion?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FVQ-Diffusion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.14822), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.16007)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fehoogeboom\u002Fmultinomial_diffusion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fimproved-diffusion)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Ws0_wK2cnsWEnfB7HtmPT4bjCPElb40C) | 30.06.2022 |\n| OPT | Open Pre-trained Transformers is a family of NLP models trained on billions of tokens of text obtained from the internet | \u003Cul>\u003Cli>[Susan Zhang](https:\u002F\u002Fgithub.com\u002Fsuchenzang)\u003C\u002Fli> \u003Cli>[Stephen Roller](https:\u002F\u002Fstephenroller.com\u002F)\u003C\u002Fli> \u003Cli>[Naman Goyal](https:\u002F\u002Fgithub.com\u002Fngoyal2707)\u003C\u002Fli> \u003Cli>[Mikel Artetxe](https:\u002F\u002Fgithub.com\u002Fartetxem)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Moya Chen](https:\u002F\u002Fmoyachen.com\u002F)\u003C\u002Fli> \u003Cli>[Christopher Dewan](https:\u002F\u002Fgithub.com\u002Fm3rlin45)\u003C\u002Fli> \u003Cli>[Mona Diab](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=-y6SIhQAAAAJ)\u003C\u002Fli> \u003Cli>[Xi Victoria Lin](http:\u002F\u002Fvictorialin.net\u002F)\u003C\u002Fli> \u003Cli>[Todor Mihaylov](https:\u002F\u002Fgithub.com\u002Ftbmihailov)\u003C\u002Fli> \u003Cli>[Myle Ott](https:\u002F\u002Fmyleott.com\u002F)\u003C\u002Fli> \u003Cli>[Sam Shleifer](https:\u002F\u002Fgithub.com\u002Fsshleifer)\u003C\u002Fli> \u003Cli>[Kurt Shuster](https:\u002F\u002Fgithub.com\u002Fklshuster)\u003C\u002Fli> \u003Cli>[Daniel Simig](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=TtWU9fsAAAAJ)\u003C\u002Fli> \u003Cli>[Punit Singh Koura](https:\u002F\u002Fgithub.com\u002Fpunitkoura)\u003C\u002Fli> \u003Cli>[Anjali Sridhar](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fanjalisridhar\u002F)\u003C\u002Fli> \u003Cli>[Tianlu Wang](https:\u002F\u002Ftianlu-wang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Luke Zettlemoyer](https:\u002F\u002Fwww.cs.washington.edu\u002Fpeople\u002Ffaculty\u002Flsz\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fmetaseq?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fmetaseq\u002Ftree\u002Fmain\u002Fprojects\u002FOPT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.01068), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.02243), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.10350), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.11990)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fai.facebook.com\u002Fblog\u002Fdemocratizing-access-to-large-scale-language-models-with-opt-175b\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FEjg0OunCi9U)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F14wnxMvD9zsiBQo2FtTpxn6w2cpXCcb-7) | 29.06.2022 |\n| Customizing a Transformer Encoder | We will learn how to customize the encoder to employ new network architectures | [Chen Chen](https:\u002F\u002Fgithub.com\u002FchenGitHuber) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Fmodels?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fmodels\u002Ftree\u002Fmaster\u002Fofficial\u002Fnlp\u002Fmodeling) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.03762)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fmodels\u002Fblob\u002Fmaster\u002Fofficial\u002Fnlp\u002Fmodeling\u002Fnetworks\u002Fencoder_scaffold.py)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fmodels\u002Fblob\u002Fmaster\u002Fofficial\u002Fcolab\u002Fnlp\u002Fcustomize_encoder.ipynb) | 22.06.2022 |\n| MTTR | End-to-End Referring Video Object Segmentation with Multimodal Transformers | \u003Cul>\u003Cli>[Adam Botach](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fadam-botach)\u003C\u002Fli> \u003Cli>[Evgenii Zheltonozhskii](https:\u002F\u002Fevgeniizh.com\u002F)\u003C\u002Fli> \u003Cli>[Chaim Baskin](https:\u002F\u002Fgithub.com\u002Fchaimbaskin)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ba7fa64376b6.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00493) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmttr2021\u002FMTTR?style=social)](https:\u002F\u002Fgithub.com\u002Fmttr2021\u002FMTTR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.14821), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.11692), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.13230)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FSwinTransformer\u002FVideo-Swin-Transformer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FMTTR\u002FMTTR-Referring-Video-Object-Segmentation)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYqlhXgq6hcs)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F12p0jpSx3pJNfZk-y_L44yeHZlhsKVra-) | 20.06.2022 |\n| SwinIR | Image Restoration Using Swin Transformer | \u003Cul>\u003Cli>[Jingyun Liang](https:\u002F\u002Fjingyunliang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiezhang Cao](https:\u002F\u002Fgithub.com\u002Fcaojiezhang)\u003C\u002Fli> \u003Cli>[Guolei Sun](https:\u002F\u002Fgithub.com\u002FGuoleiSun)\u003C\u002Fli> \u003Cli>[Kai Zhang](https:\u002F\u002Fcszn.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Luc Van Gool](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=TwMib_QAAAAJ)\u003C\u002Fli> \u003Cli>[Radu Timofte](https:\u002F\u002Fwww.informatik.uni-wuerzburg.de\u002Fcomputervision\u002Fhome\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_59e73815c1c5.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCVW54120.2021.00210) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FJingyunLiang\u002FSwinIR?style=social)](https:\u002F\u002Fgithub.com\u002FJingyunLiang\u002FSwinIR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.10257), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.10833)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcszn\u002FBSRGAN), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FSwin-Transformer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcszn\u002FKAIR)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgist\u002FJingyunLiang\u002Fa5e3e54bc9ef8d7bf594f6fee8208533\u002Fswinir-demo-on-real-world-image-sr.ipynb) | 17.06.2022 |\n| VRT | A Video Restoration Transformer | \u003Cul>\u003Cli>[Jingyun Liang](https:\u002F\u002Fjingyunliang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiezhang Cao](https:\u002F\u002Fgithub.com\u002Fcaojiezhang)\u003C\u002Fli> \u003Cli>[Yuchen Fan](https:\u002F\u002Fychfan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Kai Zhang](https:\u002F\u002Fcszn.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Yawei Li](https:\u002F\u002Fofsoundof.github.io\u002F)\u003C\u002Fli> \u003Cli>[Radu Timofte](https:\u002F\u002Fwww.informatik.uni-wuerzburg.de\u002Fcomputervision\u002Fhome\u002F)\u003C\u002Fli> \u003Cli>[Luc Van Gool](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=TwMib_QAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_aedd82489829.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTIP.2024.3372454) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FJingyunLiang\u002FVRT?style=social)](https:\u002F\u002Fgithub.com\u002FJingyunLiang\u002FVRT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.12288)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcszn\u002FKAIR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FSwinTransformer\u002FVideo-Swin-Transformer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopen-mmlab\u002Fmmediting)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgist\u002FJingyunLiang\u002Fdeb335792768ad9eb73854a8efca4fe0\u002Fvrt-demo-on-video-restoration.ipynb) | 15.06.2022 |\n| Omnivore | A single model which excels at classifying images, videos, and single-view 3D data using exactly the same model parameters | \u003Cul>\u003Cli>[Rohit Girdhar](http:\u002F\u002Frohitgirdhar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mannat Singh](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QOO8OCcAAAAJ)\u003C\u002Fli> \u003Cli>[Nikhila Ravi](https:\u002F\u002Fnikhilaravi.com\u002F)\u003C\u002Fli> \u003Cli>[Laurens Maaten](https:\u002F\u002Flvdmaaten.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Armand Joulin](https:\u002F\u002Fai.facebook.com\u002Fpeople\u002Farmand-joulin\u002F)\u003C\u002Fli> \u003Cli>[Ishan Misra](https:\u002F\u002Fimisra.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_51a58de9ffd7.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01563) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fomnivore?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fomnivore) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.08377), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2206.08356)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002Fomnivore)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ffacebookresearch.github.io\u002Fomnivore\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fdataset\u002Fepic-kitchens-100)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fomnivore\u002Fblob\u002Fmain\u002Finference_tutorial.ipynb) | 14.06.2022 |\n| Dream Fields | Zero-Shot Text-Guided Object Generation | \u003Cul>\u003Cli>[Ajay Jain](https:\u002F\u002Fajayj.com\u002F)\u003C\u002Fli> \u003Cli>[Ben Mildenhall](https:\u002F\u002Fbmild.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jon Barron](https:\u002F\u002Fjonbarron.info\u002F)\u003C\u002Fli> \u003Cli>[Pieter Abbeel](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~pabbeel\u002F)\u003C\u002Fli> \u003Cli>[Ben Poole](https:\u002F\u002Fcs.stanford.edu\u002F~poole\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_d53812bab803.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00094) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fgoogle-research?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fgoogle-research\u002Ftree\u002Fmaster\u002Fdreamfields) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.01455), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.00677), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.13415)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fajayjain\u002FDietNeRF), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fmipnerf)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fajayj.com\u002Fdreamfields)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F1Fke6w46tv4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F17GtPqdUCbG5CsmTnQFecPpoq_zpNKX7A) | 10.06.2022 |\n| Detic | Detecting Twenty-thousand Classes using Image-level Supervision | \u003Cul>\u003Cli>[Xingyi Zhou](https:\u002F\u002Fwww.cs.utexas.edu\u002F~zhouxy\u002F)\u003C\u002Fli> \u003Cli>[Rohit Girdhar](https:\u002F\u002Frohitgirdhar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Armand Joulin](https:\u002F\u002Fai.facebook.com\u002Fpeople\u002Farmand-joulin\u002F)\u003C\u002Fli> \u003Cli>[Philipp Krähenbühl](https:\u002F\u002Fgithub.com\u002Fphilkr)\u003C\u002Fli> \u003Cli>[Ishan Misra](https:\u002F\u002Fimisra.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_035daec47c0f.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20077-9_21) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002FDetic?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FDetic) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.02605)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flvis-dataset\u002Flvis-api)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1QtTW9-ukX2HKZGvt0QvVGqjuqEykoZKI) | 07.06.2022 |\n| SimCTG | Contrastive training objective to calibrate the model's representation space, and a decoding method -- contrastive search -- to encourage diversity while maintaining coherence in the generated text | \u003Cul>\u003Cli>[Yixuan Su](https:\u002F\u002Fyxuansu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tian Lan](https:\u002F\u002Fgithub.com\u002FgmftbyGMFTBY)\u003C\u002Fli> \u003Cli>[Yan Wang](https:\u002F\u002Flibertywing.github.io\u002Fyanwang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Dani Yogatama](https:\u002F\u002Fdyogatama.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lingpeng Kong](https:\u002F\u002Fikekonglp.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nigel Collier](https:\u002F\u002Fgithub.com\u002Fnhcollier)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyxuansu\u002FSimCTG?style=social)](https:\u002F\u002Fgithub.com\u002Fyxuansu\u002FSimCTG) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2202.06417), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2210.14140)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyxuansu\u002FContrastive_Search_versus_Contrastive_Decoding), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyxuansu\u002FContrastive_Search_Is_What_You_Need)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Fintroducing-csearch), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fjoaogante\u002Fcontrastive_search_generation), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Ftransformers\u002Fmodel_doc\u002Fopt)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2022\u002Fhash\u002F871cae8f599cb8bbfcb0f58fe1af95ad-Abstract-Conference.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fsimctg\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1ImvR-ldHf9rJyFzOCMJ_zjAGK8n1iTW7) | 04.06.2022 |\n| T0 | Multitask Prompted Training Enables Zero-Shot Task Generalization | \u003Cul>\u003Cli>[Victor Sanh](https:\u002F\u002Fgithub.com\u002FVictorSanh)\u003C\u002Fli> \u003Cli>[Albert Webson](https:\u002F\u002Frepresentation.ai\u002F)\u003C\u002Fli> \u003Cli>[Colin Raffel](https:\u002F\u002Fcolinraffel.com\u002F\u002F)\u003C\u002Fli> \u003Cli>[Stephen Bach](http:\u002F\u002Fcs.brown.edu\u002Fpeople\u002Fsbach\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lintang Sutawika](https:\u002F\u002Fgithub.com\u002Flintangsutawika)\u003C\u002Fli> \u003Cli>[Zaid Alyafeai](https:\u002F\u002Fgithub.com\u002Fzaidalyafeai)\u003C\u002Fli> \u003Cli>[Antoine Chaffin](https:\u002F\u002Fantoine.chaffin.fr\u002F)\u003C\u002Fli> \u003Cli>[Arnaud Stiegler](https:\u002F\u002Fgithub.com\u002Farnaudstiegler)\u003C\u002Fli> \u003Cli>[Teven Scao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ik0_vxsAAAAJ)\u003C\u002Fli> \u003Cli>[Arun Raja](https:\u002F\u002Fwww.arunraja.dev\u002F)\u003C\u002Fli> \u003Cli>[Manan Dey](https:\u002F\u002Fgithub.com\u002Fmanandey)\u003C\u002Fli> \u003Cli>[M Saiful Bari](https:\u002F\u002Fsbmaruf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Canwen Xu](https:\u002F\u002Fwww.canwenxu.net\u002F)\u003C\u002Fli> \u003Cli>[Urmish Thakker](https:\u002F\u002Fgithub.com\u002FUrmish)\u003C\u002Fli> \u003Cli>[Shanya Sharma](https:\u002F\u002Fshanyas10.github.io\u002F)\u003C\u002Fli> \u003Cli>[Eliza Szczechla](https:\u002F\u002Felsanns.github.io\u002F)\u003C\u002Fli> \u003Cli>[Taewoon Kim](https:\u002F\u002Ftae898.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gunjan Chhablani](https:\u002F\u002Fgchhablani.github.io\u002F)\u003C\u002Fli> \u003Cli>[Nihal Nayak](https:\u002F\u002Fnihalnayak.github.io\u002F)\u003C\u002Fli> \u003Cli>[Debajyoti Datta](http:\u002F\u002Fdebajyotidatta.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jonathan Chang](https:\u002F\u002Fgithub.com\u002Fcccntu\u002F)\u003C\u002Fli> \u003Cli>[Mike Tian-Jian Jiang](https:\u002F\u002Fgithub.com\u002Ftianjianjiang)\u003C\u002Fli> \u003Cli>[Matteo Manica](https:\u002F\u002Fgithub.com\u002Fdrugilsberg)\u003C\u002Fli> \u003Cli>[Sheng Shen](https:\u002F\u002Fsincerass.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zheng Xin Yong](https:\u002F\u002Fyongzx.github.io\u002F)\u003C\u002Fli> \u003Cli>[Harshit Pandey](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=BPIs78gAAAAJ)\u003C\u002Fli> \u003Cli>[Rachel Bawden](https:\u002F\u002Frbawden.github.io\u002F)\u003C\u002Fli> \u003Cli>[Trishala Neeraj](https:\u002F\u002Fgithub.com\u002Ftrishalaneeraj)\u003C\u002Fli> \u003Cli>[Jos Rozen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=OxEDKogAAAAJ)\u003C\u002Fli> \u003Cli>[Abheesht Sharma](https:\u002F\u002Fgithub.com\u002Fabheesht-sharma)\u003C\u002Fli> \u003Cli>[Andrea Santilli](https:\u002F\u002Fteelinsan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Thibault Fevry](http:\u002F\u002Fthibaultfevry.com\u002F)\u003C\u002Fli> \u003Cli>[Jason Alan Fries](https:\u002F\u002Fweb.stanford.edu\u002F~jfries\u002F)\u003C\u002Fli> \u003Cli>[Ryan Teehan](https:\u002F\u002Fgithub.com\u002Frteehas)\u003C\u002Fli> \u003Cli>[Stella Biderman](https:\u002F\u002Fwww.stellabiderman.com\u002F)\u003C\u002Fli> \u003Cli>[Leo Gao](https:\u002F\u002Fgithub.com\u002Fleogao2)\u003C\u002Fli> \u003Cli>[Tali Bers](https:\u002F\u002Fgithub.com\u002Ftbers-coursera)\u003C\u002Fli> \u003Cli>[Thomas Wolf](https:\u002F\u002Fthomwolf.io\u002F)\u003C\u002Fli> \u003Cli>[Alexander Rush](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=LIjnUGgAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fbigscience-workshop\u002Fpromptsource?style=social)](https:\u002F\u002Fgithub.com\u002Fbigscience-workshop\u002Fpromptsource) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2110.08207)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FiJ0IVZgGjTM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FYToXXfrIu6w)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1xx7SgdLaAu23YFBirXmaQViDr8caowX_) | 29.05.2022 |\n| AvatarCLIP | A zero-shot text-driven framework for 3D avatar generation and animation | \u003Cul>\u003Cli>[Fangzhou Hong](https:\u002F\u002Fhongfz16.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mingyuan Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=2QLD4fAAAAAJ)\u003C\u002Fli> \u003Cli>[Liang Pan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=lSDISOcAAAAJ)\u003C\u002Fli> \u003Cli>[Zhongang Cai](https:\u002F\u002Fcaizhongang.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Lei Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=jZH2IPYAAAAJ)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0d0b7192547c.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3528223.3530094) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhongfz16\u002FAvatarCLIP?style=social)](https:\u002F\u002Fgithub.com\u002Fhongfz16\u002FAvatarCLIP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.08535), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.01455), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.03221), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.05139), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.13333)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fwww.di.ens.fr\u002Fwillow\u002Fresearch\u002Fsurreal\u002Fdata\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdaniilidis-group\u002Fneural_renderer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FGuyTevet\u002FMotionCLIP), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTotoro97\u002FNeuS), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnghorbani\u002Fhuman_body_prior)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fhongfz16.github.io\u002Fprojects\u002FAvatarCLIP.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-l2ZMeoASGY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1dfaecX7xF3nP6fyXc8XBljV5QY1lc1TR) | 15.05.2022 |\n| Text2Mesh | Text-Driven Neural Stylization for Meshes | \u003Cul>\u003Cli>[Oscar Michel](https:\u002F\u002Fojmichel.github.io\u002F)\u003C\u002Fli> \u003Cli>[Roi Bar-On](https:\u002F\u002Fgithub.com\u002Froibaron)\u003C\u002Fli> \u003Cli>[Richard Liu](https:\u002F\u002Fgithub.com\u002Ffactoryofthesun)\u003C\u002Fli> \u003Cli>[Sagie Benaim](https:\u002F\u002Fsagiebenaim.github.io\u002F)\u003C\u002Fli> \u003Cli>[Rana Hanocka](http:\u002F\u002Fpeople.cs.uchicago.edu\u002F~ranahanocka\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fthreedle\u002Ftext2mesh?style=social)](https:\u002F\u002Fgithub.com\u002Fthreedle\u002Ftext2mesh) \u003Cul>\u003Cli>[CLIP](https:\u002F\u002Fopenai.com\u002Fblog\u002Fclip\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.03221)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fcode\u002Fneverix\u002Ftext2mesh\u002Fnotebook)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fthreedle.github.io\u002Ftext2mesh\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fthreedle\u002Ftext2mesh\u002Fblob\u002Fmaster\u002Fcolab_demo.ipynb) | 14.05.2022 |\n| T5 | Text-To-Text Transfer Transformer | \u003Cul>\u003Cli>[Colin Raffel](https:\u002F\u002Fcolinraffel.com\u002F)\u003C\u002Fli> \u003Cli>[Noam Shazeer](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=wsGvgA8AAAAJ)\u003C\u002Fli> \u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Katherine Lee](https:\u002F\u002Fgithub.com\u002Fkatelee168)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Sharan Narang](https:\u002F\u002Fgithub.com\u002Fsharannarang)\u003C\u002Fli> \u003Cli>[Michael Matena](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=rN_9vroAAAAJ)\u003C\u002Fli> \u003Cli>[Yanqi Zhou](https:\u002F\u002Fzhouyanqi.github.io)\u003C\u002Fli> \u003Cli>[Wei Li](https:\u002F\u002Fresearch.google\u002Fpeople\u002F106528\u002F)\u003C\u002Fli> \u003Cli>[Peter J. Liu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=1EPxhywAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Ftext-to-text-transfer-transformer?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Ftext-to-text-transfer-transformer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.10683)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fmesh\u002Ftree\u002Fmaster\u002Fmesh_tensorflow\u002Ftransformer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fdatasets)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Ftext-to-text-transfer-transformer\u002Fblob\u002Fmain\u002Fnotebooks\u002Ft5-trivia.ipynb) | 11.05.2022 |\n| XLS-R | Self-supervised Cross-lingual Speech Representation Learning at Scale | \u003Cul>\u003Cli>[Arun Babu](https:\u002F\u002Fgithub.com\u002Farbabu123)\u003C\u002Fli> \u003Cli>[Changhan Wang](https:\u002F\u002Fwww.changhan.me\u002F)\u003C\u002Fli> \u003Cli>[Andros Tjandra](https:\u002F\u002Fgithub.com\u002Fandrostj)\u003C\u002Fli> \u003Cli>[Kushal Lakhotia](https:\u002F\u002Fabout.me\u002Fhikushalhere)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Qiantong Xu](https:\u002F\u002Fgithub.com\u002Fxuqiantong)\u003C\u002Fli> \u003Cli>[Naman Goyal](https:\u002F\u002Fgithub.com\u002Fngoyal2707)\u003C\u002Fli> \u003Cli>[Kritika Singh](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Ltk3SykAAAAJ)\u003C\u002Fli> \u003Cli>[Patrick von Platen](https:\u002F\u002Fgithub.com\u002Fpatrickvonplaten)\u003C\u002Fli> \u003Cli>[Yatharth Saraf](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=KJTtNJwAAAAJ)\u003C\u002Fli> \u003Cli>[Juan Pino](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=weU_-4IAAAAJ)\u003C\u002Fli> \u003Cli>[Alexei Baevski](https:\u002F\u002Fgithub.com\u002Falexeib)\u003C\u002Fli> \u003Cli>[Alexis Conneau](https:\u002F\u002Fgithub.com\u002Faconneau)\u003C\u002Fli> \u003Cli>[Michael Auli](https:\u002F\u002Fgithub.com\u002Fmichaelauli)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Ffairseq?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffairseq\u002Fblob\u002Fmain\u002Fexamples\u002Fwav2vec\u002Fxlsr\u002FREADME.md) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.09296)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fhuggingface.co\u002Fblog\u002Ffine-tune-xlsr-wav2vec2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffairscale)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fpatrickvonplaten\u002Fnotebooks\u002Fblob\u002Fmaster\u002FFine_Tune_XLS_R_on_Common_Voice.ipynb) | 10.05.2022 |\n| MAGIC | Training-free framework, iMAge-Guided text generatIon with CLIP, for plugging in visual controls in the generation process and enabling LMs to perform multimodal tasks in a zero-shot manner | \u003Cul>\u003Cli>[Yixuan Su](https:\u002F\u002Fyxuansu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Tian Lan](https:\u002F\u002Fgithub.com\u002FgmftbyGMFTBY)\u003C\u002Fli> \u003Cli>[Yahui Liu](https:\u002F\u002Fyhlleo.github.io\u002F)\u003C\u002Fli> \u003Cli>[Fangyu Liu](https:\u002F\u002Ffangyuliu.me\u002Fabout)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dani Yogatama](https:\u002F\u002Fdyogatama.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yan Wang](https:\u002F\u002Flibertywing.github.io\u002Fyanwang.github.io\u002F)\u003C\u002Fli> \u003Cli>[Lingpeng Kong](https:\u002F\u002Fwww.cs.cmu.edu\u002F~lingpenk\u002F)\u003C\u002Fli> \u003Cli>[Nigel Collier](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fnhcollier\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyxuansu\u002Fmagic?style=social)](https:\u002F\u002Fgithub.com\u002Fyxuansu\u002Fmagic) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.02655)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1NDVkKpanbsaUwecHoRp_2kIpMztOFW25) | 02.05.2022 |\n| DiffCSE | Unsupervised contrastive learning framework for learning sentence embeddings | \u003Cul>\u003Cli>[Yung-Sung Chuang](https:\u002F\u002Fpeople.csail.mit.edu\u002Fyungsung\u002F)\u003C\u002Fli> \u003Cli>[Rumen Dangovski](http:\u002F\u002Fsuper-ms.mit.edu\u002Frumen.html)\u003C\u002Fli> \u003Cli>[Hongyin Luo](https:\u002F\u002Fluohongyin.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yang Zhang](https:\u002F\u002Fmitibmwatsonailab.mit.edu\u002Fpeople\u002Fyang-zhang\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Shiyu Chang](https:\u002F\u002Fcode-terminator.github.io\u002F)\u003C\u002Fli> \u003Cli>[Marin Soljačić](http:\u002F\u002Fwww.mit.edu\u002F~soljacic\u002Fmarin.html)\u003C\u002Fli> \u003Cli>[Shang-Wen Li](https:\u002F\u002Fswdanielli.github.io\u002F)\u003C\u002Fli> \u003Cli>[Scott Wen-tau Yih](https:\u002F\u002Fscottyih.org\u002F)\u003C\u002Fli> \u003Cli>[Yoon Kim](https:\u002F\u002Fpeople.csail.mit.edu\u002Fyoonkim\u002F)\u003C\u002Fli> \u003Cli>[James Glass](http:\u002F\u002Fgroups.csail.mit.edu\u002Fsls\u002Fpeople\u002Fglass.shtml)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fvoidism\u002Fdiffcse?style=social)](https:\u002F\u002Fgithub.com\u002Fvoidism\u002Fdiffcse) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.10298), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.08821), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.00899)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fprinceton-nlp\u002FSimCSE)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fvoidism)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002FYungSungChuang\u002Fstatus\u002F1517518077902000129)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fvoidism\u002FDiffCSE\u002Fblob\u002Fmaster\u002Fdiffcse_evaluation.ipynb) | 24.04.2022 |\n| ViDT+ | An Extendable, Efficient and Effective Transformer-based Object Detector | \u003Cul>\u003Cli>[Hwanjun Song](https:\u002F\u002Fsonghwanjun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Deqing Sun](https:\u002F\u002Fdeqings.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sanghyuk Chun](https:\u002F\u002Fsanghyukchun.github.io\u002Fhome\u002F)\u003C\u002Fli> \u003Cli>[Varun Jampani](https:\u002F\u002Fvarunjampani.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dongyoon Han](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fdyhan0920\u002F)\u003C\u002Fli> \u003Cli>[Byeongho Heo](https:\u002F\u002Fsites.google.com\u002Fview\u002Fbyeongho-heo\u002Fhome)\u003C\u002Fli> \u003Cli>[Wonjae Kim](https:\u002F\u002Fwonjae.kim\u002F)\u003C\u002Fli> \u003Cli>[Ming-Hsuan Yang](http:\u002F\u002Ffaculty.ucmerced.edu\u002Fmhyang\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fnaver-ai\u002Fvidt?style=social)](https:\u002F\u002Fgithub.com\u002Fnaver-ai\u002Fvidt\u002Ftree\u002Fvidt-plus) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.07962), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2110.03921)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffundamentalvision\u002FDeformable-DETR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEherSenaw\u002FViDT_colab)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FEherSenaw\u002FViDT_colab\u002Fblob\u002Fmain\u002Fvidt_colab.ipynb) | 20.04.2022 |\n| BasicVSR++ | Redesign BasicVSR by proposing second-order grid propagation and flow-guided deformable alignment | \u003Cul>\u003Cli>[Kelvin Chan](https:\u002F\u002Fckkelvinchan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shangchen Zhou](https:\u002F\u002Fshangchenzhou.com\u002F)\u003C\u002Fli> \u003Cli>[Xiangyu Xu](https:\u002F\u002Fxuxy09.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c2a923ac3976.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00588) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fckkelvinchan\u002FBasicVSR_PlusPlus?style=social)](https:\u002F\u002Fgithub.com\u002Fckkelvinchan\u002FBasicVSR_PlusPlus) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.13371)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fckkelvinchan\u002FBasicVSR-IconVSR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fckkelvinchan\u002Foffset-fidelity-loss)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fckkelvinchan.github.io\u002Fprojects\u002FBasicVSR++\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FiIDml09CUc4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1I0kZMM0DQyb4ueHZw5si8fMnRCJ_eUX3) | 18.04.2022 |\n| NAFNet | Nonlinear Activation Free Network for Image Restoration | \u003Cul>\u003Cli>[Liangyu Chen](https:\u002F\u002Fgithub.com\u002Fmayorx)\u003C\u002Fli> \u003Cli>[Xiaojie Chu](https:\u002F\u002Fgithub.com\u002Fchuxiaojie)\u003C\u002Fli> \u003Cli>[Xiangyu Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=yuB-cfoAAAAJ)\u003C\u002Fli> \u003Cli>[Jian Sun](http:\u002F\u002Fwww.jiansun.org\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_827efba6b6c2.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20071-7_2) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmegvii-research\u002FNAFNet?style=social)](https:\u002F\u002Fgithub.com\u002Fmegvii-research\u002FNAFNet) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.04676), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.08714)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fimage-deblurring-on-gopro?p=simple-baselines-for-image-restoration), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fimage-denoising-on-sidd?p=simple-baselines-for-image-restoration)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1dkO5AyktmBoWwxBwoKFUurIDn0m4qDXT) | 15.04.2022 |\n| Panini-Net | GAN Prior based Degradation-Aware Feature Interpolation for Face Restoration | \u003Cul>\u003Cli>[Yinhuai Wang](https:\u002F\u002Fgithub.com\u002Fwyhuai)\u003C\u002Fli> \u003Cli>[Yujie Hu](https:\u002F\u002Fvilla.jianzhang.tech\u002Fpeople\u002Fyujie-hu\u002F)\u003C\u002Fli> \u003Cli>[Jian Zhang](http:\u002F\u002Fjianzhang.tech\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_dc79c78aa8d7.png)](https:\u002F\u002Fdoi.org\u002F10.1609\u002Faaai.v36i3.20159) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjianzhangcs\u002Fpanini?style=social)](https:\u002F\u002Fgithub.com\u002Fjianzhangcs\u002Fpanini) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.08444)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftkarras\u002Fprogressive_growing_of_gans)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGeeveGeorge\u002FPanini-Net-Colab\u002Fblob\u002Fmain\u002FPaniniNet_Working.ipynb) | 13.04.2022 |\n| E2FGVI | An End-to-End framework for Flow-Guided Video Inpainting through elaborately designed three trainable modules, namely, flow completion, feature propagation, and content hallucination modules | \u003Cul>\u003Cli>[Zhen Li](https:\u002F\u002Fpaper99.github.io\u002F)\u003C\u002Fli> \u003Cli>[Cheng-Ze Lu](https:\u002F\u002Fgithub.com\u002FLGYoung)\u003C\u002Fli> \u003Cli>[Jianhua Qin](https:\u002F\u002Fscholar.google.com\u002Fcitations?&user=TAr7TU4AAAAJ)\u003C\u002Fli> \u003Cli>[Chun-Le Guo](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RZLYwR0AAAAJ)\u003C\u002Fli> \u003Cli>[Ming-Ming Cheng](https:\u002F\u002Fmmcheng.net\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9e584c39a658.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01704) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FMCG-NKU\u002FE2FGVI?style=social)](https:\u002F\u002Fgithub.com\u002FMCG-NKU\u002FE2FGVI) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.02663)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcompetitions.codalab.org\u002Fcompetitions\u002F19544#participate-get-data), [data](https:\u002F\u002Fdata.vision.ee.ethz.ch\u002Fcsergi\u002Fshare\u002Fdavis\u002FDAVIS-2017-trainval-480p.zip)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fresearchmm\u002FSTTN), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FFocal-Transformer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fruiliu-ai\u002FFuseFormer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fphoenix104104\u002Ffast_blind_video_consistency#evaluation)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fmlearning-ai\u002Fend-to-end-framework-for-flow-guided-video-inpainting-c5e2d8b61d20)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FN--qC3T2wc4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3eH3Fm6gOFk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F12rwY2gtG8jVWlNx9pjmmM8uGmh5ue18G) | 06.04.2022 |\n| LDM | High-Resolution Image Synthesis with Latent Diffusion Models | \u003Cul>\u003Cli>[Robin Rombach](https:\u002F\u002Fgithub.com\u002Frromb)\u003C\u002Fli> \u003Cli>[Andreas Blattmann](https:\u002F\u002Fgithub.com\u002Fablattmann)\u003C\u002Fli> \u003Cli>[Dominik Lorenz](https:\u002F\u002Fgithub.com\u002Fqp-qp)\u003C\u002Fli> \u003Cli>[Patrick Esser](https:\u002F\u002Fgithub.com\u002Fpesser)\u003C\u002Fli> \u003Cli>[Björn Ommer](https:\u002F\u002Fommer-lab.com\u002Fpeople\u002Fommer\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_7578c7a947c2.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01042) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCompVis\u002Flatent-diffusion?style=social)](https:\u002F\u002Fgithub.com\u002FCompVis\u002Flatent-diffusion) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10752), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2202.09778), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.02114)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffyu\u002Flsun), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fguided-diffusion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flucidrains\u002Fdenoising-diffusion-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flucidrains\u002Fx-transformers)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fmultimodalart\u002Flatentdiffusion)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FCompVis\u002Flatent-diffusion\u002Fblob\u002Fmaster\u002Fscripts\u002Flatent_imagenet_diffusion.ipynb) | 04.04.2022 |\n| GP-UNIT | Novel framework, Generative Prior-guided UNsupervised Image-to-image Translation, to improve the overall quality and applicability of the translation algorithm | \u003Cul>\u003Cli>[Shuai Yang](https:\u002F\u002Fwilliamyang1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liming Jiang](https:\u002F\u002Fliming-jiang.com\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwilliamyang1991\u002FGP-UNIT?style=social)](https:\u002F\u002Fgithub.com\u002Fwilliamyang1991\u002FGP-UNIT) \u003Cul>\u003Cli>[ImageNet](https:\u002F\u002Fimage-net.org\u002Fdownload.php)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2204.03641)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fclovaai\u002Fstargan-v2#datasets-and-pre-trained-networks), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fswitchablenorms\u002FCelebAMask-HQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fmetfaces-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTreB1eN\u002FInsightFace_Pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002FSPADE), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fnvlabs\u002Fimaginaire), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01779)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fproject\u002Fgpunit\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdDApWs_oDrM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fwilliamyang1991\u002FGP-UNIT\u002Fblob\u002Fmain\u002Fnotebooks\u002Finference_playground.ipynb) | 02.04.2022 |\n| DualStyleGAN | More challenging exemplar-based high-resolution portrait style transfer by introducing a novel DualStyleGAN with flexible control of dual styles of the original face domain and the extended artistic portrait domain | \u003Cul>\u003Cli>[Shuai Yang](https:\u002F\u002Fwilliamyang1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Liming Jiang](https:\u002F\u002Fliming-jiang.com\u002F)\u003C\u002Fli> \u003Cli>[Ziwei Liu](https:\u002F\u002Fliuziwei7.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0239aadbb33a.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00754) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwilliamyang1991\u002FDualStyleGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fwilliamyang1991\u002FDualStyleGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.13248)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcs.nju.edu.cn\u002Frl\u002FWebCaricature.htm), [data](https:\u002F\u002Fwww.gwern.net\u002FCrops#danbooru2019-portraits)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flowfuel\u002Fprogrock-stable), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTreB1eN\u002FInsightFace_Pytorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FGradio-Blocks\u002FDualStyleGAN), [\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fhysts\u002FDualStyleGAN)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fproject\u002Fdualstylegan\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FscZTu77jixI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fwilliamyang1991\u002FDualStyleGAN\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 24.03.2022 |\n| CLIPasso | Semantically-Aware Object Sketching | \u003Cul>\u003Cli>[Yael Vinker](https:\u002F\u002Fyaelvi116.wixsite.com\u002Fmysite)\u003C\u002Fli> \u003Cli>[Ehsan Pajouheshgar](https:\u002F\u002Fpajouheshgar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jessica Y. Bo](https:\u002F\u002Fjessica-bo.github.io\u002F)\u003C\u002Fli> \u003Cli>[Roman Bachmann](https:\u002F\u002Froman-bachmann.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Amit Bermano](https:\u002F\u002Fwww.cs.tau.ac.il\u002F~amberman\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli> \u003Cli>[Amir Zamir](https:\u002F\u002Fvilab.epfl.ch\u002Fzamir\u002F)\u003C\u002Fli> \u003Cli>[Ariel Shamir](https:\u002F\u002Ffaculty.runi.ac.il\u002Farik\u002Fsite\u002Findex.asp)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyael-vinker\u002FCLIPasso?style=social)](https:\u002F\u002Fgithub.com\u002Fyael-vinker\u002FCLIPasso) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2202.05822), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.14843)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Freplicate.com\u002Fyael-vinker\u002Fclipasso)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FBachiLi\u002Fdiffvg)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fclipasso.github.io\u002Fclipasso\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyael-vinker\u002FCLIPasso\u002Fblob\u002Fmain\u002FCLIPasso.ipynb) | 21.03.2022 |\n| StyleSDF | A high resolution, 3D-consistent image and shape generation technique | \u003Cul>\u003Cli>[Roy Or-El](https:\u002F\u002Fhomes.cs.washington.edu\u002F~royorel\u002F)\u003C\u002Fli> \u003Cli>[Xuan Luo](https:\u002F\u002Froxanneluo.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mengyi Shan](https:\u002F\u002Fshanmy.github.io\u002F)\u003C\u002Fli> \u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jeong Joon Park](https:\u002F\u002Fjjparkcv.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ira Kemelmacher-Shlizerman](https:\u002F\u002Fwww.irakemelmacher.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_914392516a6b.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01314) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Froyorel\u002FStyleSDF?style=social)](https:\u002F\u002Fgithub.com\u002Froyorel\u002FStyleSDF) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.11427)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyenchenlin\u002Fnerf-pytorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002FSerdarHelli\u002FStyleSDF-3D)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fstylesdf.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Froyorel\u002FStyleSDF\u002Fblob\u002Fmain\u002FStyleSDF_demo.ipynb) | 05.03.2022 |\n| Disentangled Lifespan Face Synthesis | LFS model is proposed to disentangle the key face characteristics including shape, texture and identity so that the unique shape and texture age transformations can be modeled effectively | \u003Cul>\u003Cli>[Sen He](https:\u002F\u002Fsenhe.github.io\u002F)\u003C\u002Fli> \u003Cli>[Wentong Liao](https:\u002F\u002Fwww.tnt.uni-hannover.de\u002Fen\u002Fstaff\u002Fliao\u002F)\u003C\u002Fli> \u003Cli>[Michael Yang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fmichaelyingyang\u002F)\u003C\u002Fli> \u003Cli>[Yi-Zhe Song](http:\u002F\u002Fpersonal.ee.surrey.ac.uk\u002FPersonal\u002FY.Song\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Bodo Rosenhahn](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=qq3TxtcAAAAJ)\u003C\u002Fli> \u003Cli>[Tao Xiang](http:\u002F\u002Fpersonal.ee.surrey.ac.uk\u002FPersonal\u002FT.Xiang\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FSenHe\u002FDLFS?style=social)](https:\u002F\u002Fgithub.com\u002FSenHe\u002FDLFS) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.02874)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsenhe.github.io\u002Fprojects\u002Ficcv_2021_lifespan_face\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=uklX03ns0m0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1fgVAoxCSaqPkj0rUK4RmBh7GTQRqLNpE) | 22.02.2022 |\n| ClipCap | CLIP Prefix for Image Captioning | \u003Cul>\u003Cli>[Ron Mokady](https:\u002F\u002Frmokady.github.io\u002F)\u003C\u002Fli> \u003Cli>[Amir Hertz](https:\u002F\u002Fgithub.com\u002Famirhertz)\u003C\u002Fli> \u003Cli>[Amit Bermano](https:\u002F\u002Fwww.cs.tau.ac.il\u002F~amberman\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Frmokady\u002FCLIP_prefix_caption?style=social)](https:\u002F\u002Fgithub.com\u002Frmokady\u002FCLIP_prefix_caption) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.09734)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcocodataset.org\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002FCLIP_prefix_captioning)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@uppalamukesh\u002Fclipcap-clip-prefix-for-image-captioning-3970c73573bc)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVQDrmuccWDo)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Frmokady\u002FCLIP_prefix_caption\u002Fblob\u002Fmain\u002Fnotebooks\u002Fclip_prefix_captioning_inference.ipynb#scrollTo=glBzYsgIwhwF) | 15.02.2022 |\n| ROMP | Monocular, One-stage, Regression of Multiple 3D People | \u003Cul>\u003Cli>[Yu Sun](https:\u002F\u002Fwww.yusun.work\u002F)\u003C\u002Fli> \u003Cli>[Qian Bao](https:\u002F\u002Fgithub.com\u002Ffor-code0216)\u003C\u002Fli> \u003Cli>[Wu Liu](https:\u002F\u002Ffaculty.ustc.edu.cn\u002Fliuwu)\u003C\u002Fli> \u003Cli>[Yili Fu](https:\u002F\u002Fieeexplore.ieee.org\u002Fauthor\u002F37286601800)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli> \u003Cli>[Tao Mei](https:\u002F\u002Ftaomei.me\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fac7d2fbad55.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.01099) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FArthur151\u002FROMP?style=social)](https:\u002F\u002Fgithub.com\u002FArthur151\u002FROMP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.12272), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.08274), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F2306.02850)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FArthur151\u002FRelative_Human), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FArthur151\u002FDynaCam), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyanchxx\u002FMoPA)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhunBPJxnyBU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQ62fj_6AxRI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fl8aLHDXWQRw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1oz9E6uIbj4udOPZvA1Zi9pFx0SWH_UXg) | 11.02.2022 |\n| Mask2Former | Masked-attention Mask Transformer for Universal Image Segmentation | \u003Cul>\u003Cli>[Bowen Cheng](https:\u002F\u002Fbowenc0221.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ishan Misra](https:\u002F\u002Fimisra.github.io\u002F)\u003C\u002Fli> \u003Cli>[Alexander Schwing](https:\u002F\u002Falexander-schwing.de\u002F)\u003C\u002Fli> \u003Cli>[Alexander Kirillov](https:\u002F\u002Falexander-kirillov.github.io\u002F)\u003C\u002Fli> \u003Cli>[Rohit Girdhar](https:\u002F\u002Frohitgirdhar.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_7b9e52c86acd.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00135) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002FMask2Former?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FMask2Former) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.01527), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10764)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Freplicate.com\u002Ffacebookresearch\u002Fmask2former)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FMaskFormer)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002FMask2Former)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fbowenc0221.github.io\u002Fmask2former\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1uIWE5KbGFSjrxey2aRd5pWkKNY1_SaNq) | 09.02.2022 |\n| BertViz | Tool that visualizes attention at multiple scales, each of which provides a unique perspective on the attention mechanism | [Jesse Vig](https:\u002F\u002Fjessevig.com\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_28aa197db431.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002FP19-3007) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjessevig\u002Fbertviz?style=social)](https:\u002F\u002Fgithub.com\u002Fjessevig\u002Fbertviz) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.05714), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.11218), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.10186), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.04626), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.05607)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@GaryFr0sty\u002Fvisualize-attention-scores-of-llms-with-bertviz-3deb94b455b3)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fbertviz\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FumErGRrfSk4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1hXIQ77A4TYS4y3UthWF-Ci7V7vVUoxmQ) | 05.02.2022 |\n| JoJoGAN | One Shot Face Stylization | \u003Cul>\u003Cli>[Min Jin Chong](https:\u002F\u002Fmchong6.github.io\u002F)\u003C\u002Fli> \u003Cli>[David Forsyth](http:\u002F\u002Fluthuli.cs.uiuc.edu\u002F~daf\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_b840583229af.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-19787-1_8) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmchong6\u002FJoJoGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fmchong6\u002FJoJoGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.11641)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Freplicate\u002Fcog)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmchong6\u002FJoJoGAN\u002Fblob\u002Fmaster\u002Fstylize.ipynb) | 02.02.2022 |\n| Pose with Style | Detail-Preserving Pose-Guided Image Synthesis with Conditional StyleGAN | \u003Cul>\u003Cli>[Badour AlBahar](https:\u002F\u002Fbadouralbahar.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jingwan Lu](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Fjingwan-lu\u002F)\u003C\u002Fli> \u003Cli>[Jimei Yang](https:\u002F\u002Fgithub.com\u002Fjimeiyang)\u003C\u002Fli> \u003Cli>[Zhixin Shu](https:\u002F\u002Fzhixinshu.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli> \u003Cli>[Jia-Bin Huang](https:\u002F\u002Fjbhuang0604.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FBadourAlBahar\u002Fpose-with-style?style=social)](https:\u002F\u002Fgithub.com\u002FBadourAlBahar\u002Fpose-with-style) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.06166)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpose-with-style.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fd_ETeAVLilw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftg-bomze\u002Fcollection-of-notebooks\u002Fblob\u002Fmaster\u002FHomeStylist.ipynb) | 19.01.2022 |\n| ConvNeXt | A pure ConvNet model constructed entirely from standard ConvNet modules | \u003Cul>\u003Cli>[Zhuang Liu](https:\u002F\u002Fliuzhuang13.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hanzi Mao](https:\u002F\u002Fhanzimao.me\u002F)\u003C\u002Fli> \u003Cli>[Chao-Yuan Wu](https:\u002F\u002Fchaoyuan.org\u002F)\u003C\u002Fli> \u003Cli>[Christoph Feichtenhofer](https:\u002F\u002Ffeichtenhofer.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Trevor Darrell](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~trevor\u002F)\u003C\u002Fli> \u003Cli>[Saining Xie](https:\u002F\u002Fwww.sainingxie.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_5f262d9dec0d.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01167) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002FConvNeXt?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FConvNeXt) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.03545)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frwightman\u002Fpytorch-image-models), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fdeit)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002Fconvnext)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQzCjXqFnWPE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FidiIllIQOfU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQqejV0LNDHA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1CBYTIZ4tBMsVL5cqu9N_-Q3TBprqsfEO) | 19.01.2022 |\n| diffsort | Differentiable Sorting Networks | \u003Cul>\u003Cli>[Felix Petersen](http:\u002F\u002Fpetersen.ai\u002F)\u003C\u002Fli> \u003Cli>[Christian Borgelt](https:\u002F\u002Fborgelt.net\u002F)\u003C\u002Fli> \u003Cli>[Hilde Kuehne](https:\u002F\u002Fhildekuehne.github.io\u002F)\u003C\u002Fli> \u003Cli>[Oliver Deussen](https:\u002F\u002Fwww.cgmi.uni-konstanz.de\u002Fpersonen\u002Fprof-dr-oliver-deussen\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FFelix-Petersen\u002Fdiffsort?style=social)](https:\u002F\u002Fgithub.com\u002FFelix-Petersen\u002Fdiffsort) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.04019), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2203.09630)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRl-sFaE1z4M)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1q0TZFFYB9FlOJYWKt0_7ZaXQT190anhm) | 17.01.2022 |\n| Taming Transformers for High-Resolution Image Synthesis | We combine the efficiancy of convolutional approaches with the expressivity of transformers by introducing a convolutional VQGAN, which learns a codebook of context-rich visual parts, whose composition is modeled with an autoregressive transformer | \u003Cul>\u003Cli>[Patrick Esser](https:\u002F\u002Fgithub.com\u002Fpesser)\u003C\u002Fli> \u003Cli>[Robin Rombach](https:\u002F\u002Fgithub.com\u002Frromb)\u003C\u002Fli> \u003Cli>[Björn Ommer](https:\u002F\u002Fommer-lab.com\u002Fpeople\u002Fommer\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_df248106cc28.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01268) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCompVis\u002Ftaming-transformers?style=social)](https:\u002F\u002Fgithub.com\u002FCompVis\u002Ftaming-transformers) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.09841)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fcompvis.github.io\u002Ftaming-transformers\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FCompVis\u002Ftaming-transformers\u002Fblob\u002Fmaster\u002Fscripts\u002Ftaming-transformers.ipynb) | 13.01.2022 |\n| GFM | Glance and Focus Matting network, which employs a shared encoder and two separate decoders to learn both tasks in a collaborative manner for end-to-end natural image matting | \u003Cul>\u003Cli>[Jizhizi Li](https:\u002F\u002Fjizhizili.github.io\u002Fhomepage\u002F)\u003C\u002Fli> \u003Cli>[Jing Zhang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=9jH5v74AAAAJ)\u003C\u002Fli> \u003Cli>[Stephen Maybank](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=gpyHJmcAAAAJ)\u003C\u002Fli> \u003Cli>[Dacheng Tao](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=RwlJNLcAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FJizhiziLi\u002FGFM?style=social)](https:\u002F\u002Fgithub.com\u002FJizhiziLi\u002FGFM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.16188)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJizhiziLi\u002FRIM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJizhiziLi\u002FP3M), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJizhiziLi\u002FAIM), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FViTAE-Transformer\u002FP3M-Net), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FjizhiziLi\u002Fmatting-survey)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FFJPm4YQOEyo)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1EaQ5h4u9Q_MmDSFTDmFG0ZOeSsFuRTsJ) | 05.01.2022 |\n| RealBasicVSR | Investigating Tradeoffs in Real-World Video Super-Resolution | \u003Cul>\u003Cli>[Kelvin Chan](https:\u002F\u002Fckkelvinchan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shangchen Zhou](https:\u002F\u002Fshangchenzhou.com\u002F)\u003C\u002Fli> \u003Cli>[Xiangyu Xu](https:\u002F\u002Fxuxy09.github.io\u002F)\u003C\u002Fli> \u003Cli>[Chen Change Loy](https:\u002F\u002Fwww.mmlab-ntu.com\u002Fperson\u002Fccloy\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_137a3122300c.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.00587) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fckkelvinchan\u002FRealBasicVSR?style=social)](https:\u002F\u002Fgithub.com\u002Fckkelvinchan\u002FRealBasicVSR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.12704)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002FRealBasicVSR)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Ftc8p70\u002Frp_investigating_tradeoffs_in_realworld_video\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1JzWRUR34hpKvtCHm84IGx6nv35LCv20J) | 25.12.2021 |\n| GLIDE | Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models | \u003Cul>\u003Cli>[Alex Nichol](https:\u002F\u002Faqnichol.com\u002F)\u003C\u002Fli> \u003Cli>[Prafulla Dhariwal](https:\u002F\u002Fgithub.com\u002Fprafullasd)\u003C\u002Fli> \u003Cli>[Aditya Ramesh](http:\u002F\u002Fadityaramesh.com\u002F)\u003C\u002Fli> \u003Cli>[Pranav Shyam](https:\u002F\u002Fgithub.com\u002Fpranv)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Pamela Mishkin](https:\u002F\u002Fmanlikemishap.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bob McGrew](https:\u002F\u002Fgithub.com\u002Fbmcgrew)\u003C\u002Fli> \u003Cli>[Ilya Sutskever](http:\u002F\u002Fwww.cs.utoronto.ca\u002F~ilya\u002F)\u003C\u002Fli> \u003Cli>[Mark Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=5fU-QMwAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fopenai\u002Fglide-text2im?style=social)](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fglide-text2im) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2112.10741)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FItKi3h7IY2o)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fopenai\u002Fglide-text2im\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finpaint.ipynb) | 22.12.2021 |\n| Nerfies | First method capable of photorealistically reconstructing deformable scenes using photos\u002Fvideos captured casually from mobile phones | \u003Cul>\u003Cli>[Keunhong Park](https:\u002F\u002Fkeunhong.com\u002F)\u003C\u002Fli> \u003Cli>[Utkarsh Sinha](https:\u002F\u002Futkarshsinha.com\u002F)\u003C\u002Fli> \u003Cli>[Jon Barron](https:\u002F\u002Fjonbarron.info\u002F)\u003C\u002Fli> \u003Cli>[Sofien Bouaziz](http:\u002F\u002Fsofienbouaziz.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dan Goldman](https:\u002F\u002Fwww.danbgoldman.com\u002Fhome\u002F)\u003C\u002Fli> \u003Cli>[Steve Seitz](https:\u002F\u002Fwww.smseitz.com\u002F)\u003C\u002Fli> \u003Cli>[Ricardo Martin-Brualla](https:\u002F\u002Fricardomartinbrualla.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_20633cc40616.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00581) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fnerfies?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fnerfies) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.12948)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fgoogle-research\u002Ftree\u002Fmaster\u002Fjaxnerf)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnerfies.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fphotogrammetry\u002Fcomments\u002Fk1i0ct\u002Fdeformable_neural_radiance_fields_nerfies\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FMrKrnHhk8IA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIDMiMKWucaI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fnerfies\u002Fblob\u002Fmain\u002Fnotebooks\u002FNerfies_Capture_Processing.ipynb) | 06.12.2021 |\n| HyperStyle | A hypernetwork that learns to modulate StyleGAN's weights to faithfully express a given image in editable regions of the latent space | \u003Cul>\u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Omer Tov](https:\u002F\u002Fgithub.com\u002Fomertov)\u003C\u002Fli> \u003Cli>[Ron Mokady](https:\u002F\u002Frmokady.github.io\u002F)\u003C\u002Fli> \u003Cli>[Rinon Gal](https:\u002F\u002Frinongal.github.io\u002F)\u003C\u002Fli> \u003Cli>[Amit Bermano](https:\u002F\u002Fwww.cs.tau.ac.il\u002F~amberman\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_27a55ce01ec1.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR52688.2022.01796) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyuval-alaluf\u002Fhyperstyle?style=social)](https:\u002F\u002Fgithub.com\u002Fyuval-alaluf\u002Fhyperstyle) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.15666), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.03189), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.09036), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.07727)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fai.stanford.edu\u002F~jkrause\u002Fcars\u002Fcar_dataset.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fclovaai\u002Fstargan-v2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTreB1eN\u002FInsightFace_Pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHuangYG123\u002FCurricularFace), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flessw2020\u002FRanger-Deep-Learning-Optimizer), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fpytorch\u002Fvision\u002Fblob\u002Fmain\u002Ftorchvision\u002Fmodels\u002Fresnet.py), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdvschultz\u002Fstylegan2-ada-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyuval-alaluf.github.io\u002Fhyperstyle\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_sbXmLY2jMw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyuval-alaluf\u002Fhyperstyle\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 03.12.2021 |\n| encoder4editing | Designing an Encoder for StyleGAN Image Manipulation | \u003Cul>\u003Cli>[Omer Tov](https:\u002F\u002Fgithub.com\u002Fomertov)\u003C\u002Fli> \u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yotam Nitzan](https:\u002F\u002Fyotamnitzan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1a4c1c558661.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3450626.3459838) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fomertov\u002Fencoder4editing?style=social)](https:\u002F\u002Fgithub.com\u002Fomertov\u002Fencoder4editing) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.02766)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Feladrich\u002Fpixel2style2pixel)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fomertov\u002Fencoder4editing\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 02.12.2021 |\n| StyleCariGAN | Caricature Generation via StyleGAN Feature Map Modulation | \u003Cul>\u003Cli>[Wonjong Jang](https:\u002F\u002Fwonjongg.github.io\u002F)\u003C\u002Fli> \u003Cli>[Gwangjin Ju](https:\u002F\u002Fgithub.com\u002Fjugwangjin)\u003C\u002Fli> \u003Cli>[Yucheol Jung](https:\u002F\u002Fycjung.info\u002F)\u003C\u002Fli> \u003Cli>[Jiaolong Yang](https:\u002F\u002Fjlyang.org\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Xin Tong](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Fxtong\u002F)\u003C\u002Fli> \u003Cli>[Seungyong Lee](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=yGPH-nAAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9efa6c5182cc.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3450626.3459860) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fwonjongg\u002FStyleCariGAN?style=social)](https:\u002F\u002Fgithub.com\u002Fwonjongg\u002FStyleCariGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.04331)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwonjongg.github.io\u002FStyleCariGAN\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=kpHbGOlI-BU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1HDRQGm7pvC9mAb6Lktoft_SmY9sCq_Qg) | 30.11.2021 |\n| CartoonGAN | The implementation of the cartoon GAN model with PyTorch | [Tobias Sunderdiek](https:\u002F\u002Fgithub.com\u002FTobiasSunderdiek) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_f01de748abab.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR.2018.00986) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Falamson\u002Fsafebooru)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftobiassunderdiek.github.io\u002Fcartoon-gan\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FTobiasSunderdiek\u002Fcartoon-gan\u002Fblob\u002Fmaster\u002FCartoonGAN.ipynb) | 24.11.2021 |\n| SimSwap | An efficient framework, called Simple Swap, aiming for generalized and high fidelity face swapping | \u003Cul>\u003Cli>[Xuanhong Chen](https:\u002F\u002Fgithub.com\u002Fneuralchen)\u003C\u002Fli> \u003Cli>[Bingbing Ni](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=eUbmKwYAAAAJ)\u003C\u002Fli> \u003Cli>[Yanhao Ge](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=h6tuBAcAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_125d6c68e8af.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3394171.3413630) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fneuralchen\u002FSimSwap?style=social)](https:\u002F\u002Fgithub.com\u002Fneuralchen\u002FSimSwap) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.06340)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fneuralchen\u002FSimSwap\u002Fblob\u002Fmaster\u002FSimSwap%20colab.ipynb) | 24.11.2021 |\n| RVM | Robust High-Resolution Video Matting with Temporal Guidance | \u003Cul>\u003Cli>[Shanchuan Lin](https:\u002F\u002Fgithub.com\u002FPeterL1n)\u003C\u002Fli> \u003Cli>[Linjie Yang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Flinjieyang89\u002F)\u003C\u002Fli> \u003Cli>[Imran Saleemi](http:\u002F\u002Fwww.cs.ucf.edu\u002F~imran\u002F)\u003C\u002Fli> \u003Cli>[Soumyadip Sengupta](https:\u002F\u002Fhomes.cs.washington.edu\u002F~soumya91\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a1152adfc4dc.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FWACV51458.2022.00319) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPeterL1n\u002FRobustVideoMatting?style=social)](https:\u002F\u002Fgithub.com\u002FPeterL1n\u002FRobustVideoMatting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F2108.11515)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FVideoProcessingFramework), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFeiGeChuanShu\u002Fncnn_Android_RobustVideoMatting)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpeterl1n.github.io\u002FRobustVideoMatting)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJvzltozpbpk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAy-mGCEYEzM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F10z-pNKRnVNsp0Lq9tH1J_XPZ7CBC_uHm) | 24.11.2021 |\n| RVM | Robust, real-time, high-resolution human video matting method that achieves new state-of-the-art performance | \u003Cul>\u003Cli>[Shanchuan Lin](https:\u002F\u002Fgithub.com\u002FPeterL1n)\u003C\u002Fli> \u003Cli>[Linjie Yang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Flinjieyang89)\u003C\u002Fli> \u003Cli>[Imran Saleemi](https:\u002F\u002Fgithub.com\u002Fimran-saleemi)\u003C\u002Fli> \u003Cli>[Soumyadip Sengupta](https:\u002F\u002Fgithub.com\u002Fsenguptaumd)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a1152adfc4dc.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FWACV51458.2022.00319) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FPeterL1n\u002FRobustVideoMatting?style=social)](https:\u002F\u002Fgithub.com\u002FPeterL1n\u002FRobustVideoMatting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2108.11515)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpeterl1n.github.io\u002FRobustVideoMatting)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Fpdbpmg\u002Fr_robust_highresolution_video_matting_with\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJvzltozpbpk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAy-mGCEYEzM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVL-0K6HjhvQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FJhuf6M_VrBI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_oN9yyRi3HY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F10z-pNKRnVNsp0Lq9tH1J_XPZ7CBC_uHm) | 24.11.2021 |\n| AnimeGANv2 | An improved version of AnimeGAN - it prevents the generation of high-frequency artifacts by simply changing the normalization of features in the network | \u003Cul>\u003Cli>[Xin Chen](https:\u002F\u002Fgithub.com\u002FTachibanaYoshino)\u003C\u002Fli> \u003Cli>[Gang Liu](https:\u002F\u002Fgithub.com\u002Flg0061408)\u003C\u002Fli> \u003Cli>[bryandlee](https:\u002F\u002Fgithub.com\u002Fbryandlee)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c49bddb96517.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-981-15-5577-0_18) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fbryandlee\u002Fanimegan2-pytorch?style=social)](https:\u002F\u002Fgithub.com\u002Fbryandlee\u002Fanimegan2-pytorch) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTachibanaYoshino\u002FAnimeGANv2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTachibanaYoshino\u002FAnimeGAN)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002FAnimeGANv2)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Ftachibanayoshino.github.io\u002FAnimeGANv2\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fbryandlee\u002Fanimegan2-pytorch\u002Fblob\u002Fmaster\u002Fcolab_demo.ipynb) | 17.11.2021 |\n| SOAT | StyleGAN of All Trades: Image Manipulation with Only Pretrained StyleGAN | \u003Cul>\u003Cli>[Min Jin Chong](https:\u002F\u002Fmchong6.github.io\u002F)\u003C\u002Fli> \u003Cli>[Hsin-Ying Lee](http:\u002F\u002Fhsinyinglee.com\u002F)\u003C\u002Fli> \u003Cli>[David Forsyth](http:\u002F\u002Fluthuli.cs.uiuc.edu\u002F~daf\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmchong6\u002FSOAT?style=social)](https:\u002F\u002Fgithub.com\u002Fmchong6\u002FSOAT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2111.01619)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjustinpinkney\u002Ftoonify), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fakhaliq\u002FSOAT)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmchong6\u002FSOAT\u002Fblob\u002Fmaster\u002Finfinity.ipynb) | 13.11.2021 |\n| Arnheim | Generative Art Using Neural Visual Grammars and Dual Encoders | \u003Cul>\u003Cli>[Chrisantha Fernando](https:\u002F\u002Fwww.chrisantha.co.uk\u002F)\u003C\u002Fli> \u003Cli>[Ali Eslami](http:\u002F\u002Farkitus.com\u002F)\u003C\u002Fli> \u003Cli>[Jean-Baptiste Alayrac](https:\u002F\u002Fwww.jbalayrac.com\u002F)\u003C\u002Fli> \u003Cli>[Piotr Mirowski](https:\u002F\u002Fpiotrmirowski.com\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dylan Banarse](https:\u002F\u002Fwww.2ne1.com\u002F)\u003C\u002Fli> \u003Cli>[Simon Osindero](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Jq8ZS5kAAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepmind\u002Farnheim?style=social)](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Farnheim) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.00162), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.14843), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.07729), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1606.02580), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1609.09106)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fdall-e)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FCompositional_pattern-producing_network)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=U7guaMdeF4g), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zh0goLbS-l0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=SYJGNt7yu6M), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=MxkYKa0x5AU)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Farnheim\u002Fblob\u002Fmaster\u002Farnheim_2.ipynb) | 11.11.2021 |\n| StyleGAN 2 | Generation of faces, cars, etc. | [Mikael Christensen](https:\u002F\u002Fgithub.com\u002FSyntopia) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ada1741909ed.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00813) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNVlabs\u002Fstylegan2?style=social)](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F1912.04958)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fc-NJtV9Jvp0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1ShgW6wohEFQtqs_znMna3dzrcVoABKIH) | 05.11.2021 |\n| ByteTrack | Multi-Object Tracking by Associating Every Detection Box | \u003Cul>\u003Cli>[Yifu Zhang](https:\u002F\u002Fgithub.com\u002Fifzhang)\u003C\u002Fli> \u003Cli>[Peize Sun](https:\u002F\u002Fpeizesun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yi Jiang](https:\u002F\u002Fgithub.com\u002FiFighting)\u003C\u002Fli> \u003Cli>[Dongdong Yu](https:\u002F\u002Fmiracle-fmh.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ping Luo](http:\u002F\u002Fluoping.me\u002F)\u003C\u002Fli> \u003Cli>[Xinggang Wang](https:\u002F\u002Fxinggangw.info\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_cf78940e823f.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-031-20047-2_1) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fifzhang\u002FByteTrack?style=social)](https:\u002F\u002Fgithub.com\u002Fifzhang\u002FByteTrack) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2110.06864)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fmotchallenge.net\u002F), [data](https:\u002F\u002Fwww.crowdhuman.org\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMegvii-BaseDetection\u002FYOLOX), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fifzhang\u002FFairMOT), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FPeizeSun\u002FTransTrack), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsamylee\u002FTowards-Realtime-MOT-Cpp)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Ftask\u002Fmulti-object-tracking)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1bDilg4cmXFa8HCKHbsZ_p16p0vrhLyu0) | 30.10.2021 |\n| GPT-2 | Retrain an advanced text generating neural network on any text dataset using gpt-2-simple! | [Max Woolf](https:\u002F\u002Fminimaxir.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fopenai\u002Fgpt-2?style=social)](https:\u002F\u002Fgithub.com\u002Fopenai\u002Fgpt-2) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fminimaxir.com\u002F2019\u002F09\u002Fhowto-gpt2\u002F), [blog post](https:\u002F\u002Fopenai.com\u002Fresearch\u002Fbetter-language-models)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fminimaxir\u002Fgpt-2-simple)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Faqlzde\u002Fr_openai_better_language_models_and_their\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1VLG8e7YSEwypxU-noRNhsv5dW4NfTGce) | 18.10.2021 |\n| ConvMixer | An extremely simple model that is similar in spirit to the ViT and the even-more-basic MLP-Mixer in that it operates directly on patches as input, separates the mixing of spatial and channel dimensions, and maintains equal size and resolution throughout the network | \u003Cul>\u003Cli>[Asher Trockman](http:\u002F\u002Fashertrockman.com\u002F)\u003C\u002Fli> \u003Cli>[Zico Kolter](http:\u002F\u002Fzicokolter.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Flocuslab\u002Fconvmixer?style=social)](https:\u002F\u002Fgithub.com\u002Flocuslab\u002Fconvmixer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2201.09792)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flocuslab\u002Fconvmixer-cifar10), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frwightman\u002Fpytorch-image-models)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fcodex\u002Fan-overview-on-convmixer-patches-are-all-you-need-8502a8d87011)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGl0s0GDqN3c?t=990)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Flocuslab\u002Fconvmixer\u002Fblob\u002Fmain\u002Fpytorch-image-models\u002Fnotebooks\u002FEffResNetComparison.ipynb) | 06.10.2021 |\n| IC-GAN | Instance-Conditioned GAN | \u003Cul>\u003Cli>[Arantxa Casanova](https:\u002F\u002Fgithub.com\u002FArantxaCasanova)\u003C\u002Fli> \u003Cli>[Marlène Careil](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fmarl%C3%A8ne-careil-901804155)\u003C\u002Fli> \u003Cli>[Jakob Verbeek](http:\u002F\u002Fthoth.inrialpes.fr\u002F~verbeek\u002F)\u003C\u002Fli> \u003Cli>[Michał Drożdżal](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=XK_ktwQAAAAJ)\u003C\u002Fli> \u003Cli>[Adriana Romero-Soriano](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fadriromsor)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fic_gan?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fic_gan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.05070)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fai.facebook.com\u002Fblog\u002Finstance-conditioned-gans\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Ffaiss), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fajbrock\u002FBigGAN-PyTorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fbioinf-jku\u002FTTUR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fdata-efficient-gans)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2021\u002Fhash\u002Fe7ac288b0f2d41445904d071ba37aaff-Abstract.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fic_gan\u002Fblob\u002Fmaster\u002Finference\u002Ficgan_colab.ipynb) | 01.10.2021 |\n| Skillful Precipitation Nowcasting Using Deep Generative Models of Radar | Open-sourced dataset and model snapshot for precipitation nowcasting | \u003Cul>\u003Cli>[Suman Ravuri](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsuman-ravuri-81928082)\u003C\u002Fli> \u003Cli>[Karel Lenc](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~karel\u002F)\u003C\u002Fli> \u003Cli>[Matthew Willson](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fmatthew-willson-6a1b422)\u003C\u002Fli> \u003Cli>[Dmitry Kangin](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vv-leaMAAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Rémi Lam](https:\u002F\u002Fgithub.com\u002Fremilam)\u003C\u002Fli> \u003Cli>[Piotr Mirowski](https:\u002F\u002Fpiotrmirowski.com\u002F)\u003C\u002Fli> \u003Cli>[Maria Athanassiadou](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=VtkgHP0AAAAJ)\u003C\u002Fli> \u003Cli>[Sheleem Kashem](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsheleemkashem\u002F)\u003C\u002Fli> \u003Cli>[Rachel Prudden](https:\u002F\u002Fcomputerscience.exeter.ac.uk\u002Fstaff\u002Frep218)\u003C\u002Fli> \u003Cli>[Amol Mandhane](https:\u002F\u002Fgithub.com\u002Famol-mandhane)\u003C\u002Fli> \u003Cli>[Aidan Clark](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=_19DrfIAAAAJ)\u003C\u002Fli> \u003Cli>[Andrew Brock](https:\u002F\u002Fgithub.com\u002Fajbrock)\u003C\u002Fli> \u003Cli>[Karen Simonyan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=L7lMQkQAAAAJ)\u003C\u002Fli> \u003Cli>[Raia Hadsell](https:\u002F\u002Fgithub.com\u002Fraiah)\u003C\u002Fli> \u003Cli>[Niall Robinson](https:\u002F\u002Fgithub.com\u002Fniallrobinson)\u003C\u002Fli> \u003Cli>[Ellen Clancy](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fellen-clancy-815967124)\u003C\u002Fli> \u003Cli>[Shakir Mohamed](https:\u002F\u002Fwww.shakirm.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_5fb907f53a5e.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs41586-021-03854-z) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepmind\u002Fdeepmind-research?style=social)](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fdeepmind-research\u002Ftree\u002Fmaster\u002Fnowcasting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.00954)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fdeepmind.com\u002Fblog\u002Farticle\u002Fnowcasting)\u003C\u002Fli>\u003Cli>[local kernel](https:\u002F\u002Fresearch.google.com\u002Fcolaboratory\u002Flocal-runtimes.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fhub)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Fdeepmind-research\u002Fblob\u002Fmaster\u002Fnowcasting\u002FOpen_sourced_dataset_and_model_snapshot_for_precipitation_nowcasting.ipynb) | 29.09.2021 |\n| Live Speech Portraits | Real-Time Photorealistic Talking-Head Animation | \u003Cul>\u003Cli>[Yuanxun Lu](https:\u002F\u002Fgithub.com\u002FYuanxunLu)\u003C\u002Fli> \u003Cli>[Jinxiang Chai](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=OcN1_gwAAAAJ)\u003C\u002Fli> \u003Cli>[Xun Cao](https:\u002F\u002Fcite.nju.edu.cn\u002FPeople\u002FFaculty\u002F20190621\u002Fi5054.html)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c42b73ecb91d.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3478513.3480484) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYuanxunLu\u002FLiveSpeechPortraits?style=social)](https:\u002F\u002Fgithub.com\u002FYuanxunLu\u002FLiveSpeechPortraits) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.10595)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flelechen63\u002FATVGnet), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flelechen63\u002FTalking-head-Generation-with-Rhythmic-Head-Motion), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FDinoMan\u002Fspeech-driven-animation), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjunyanz\u002Fpytorch-CycleGAN-and-pix2pix)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyuanxunlu.github.io\u002Fprojects\u002FLiveSpeechPortraits\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1tKvi-9kY3GkEK8lgtfTSM70rMFo_TY50) | 26.09.2021 |\n| StylEx | Training a GAN to explain a classifier in StyleSpace | \u003Cul>\u003Cli>[Oran Lang](https:\u002F\u002Fresearch.google\u002Fpeople\u002F105975\u002F)\u003C\u002Fli> \u003Cli>[Yossi Gandelsman](https:\u002F\u002Fyossigandelsman.github.io\u002F)\u003C\u002Fli> \u003Cli>[Michal Yarom](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=GMVxiYgAAAAJ)\u003C\u002Fli> \u003Cli>[Yoav Wald](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=hh5nOn4AAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Gal Elidan](https:\u002F\u002Fresearch.google\u002Fpeople\u002F105719\u002F)\u003C\u002Fli> \u003Cli>[Avinatan Hassidim](https:\u002F\u002Fresearch.google\u002Fpeople\u002F105831\u002F)\u003C\u002Fli> \u003Cli>[William Freeman](https:\u002F\u002Fbillf.mit.edu\u002F)\u003C\u002Fli> \u003Cli>[Phillip Isola](http:\u002F\u002Fweb.mit.edu\u002Fphillipi\u002F)\u003C\u002Fli> \u003Cli>[Amir Globerso](https:\u002F\u002Fcs3801.wixsite.com\u002Famirgloberson)\u003C\u002Fli> \u003Cli>[Michal Irani](http:\u002F\u002Fwww.weizmann.ac.il\u002Fmath\u002Firani\u002F)\u003C\u002Fli> \u003Cli>[Inbar Mosseri](https:\u002F\u002Fresearch.google\u002Fpeople\u002FInbarMosseri\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_50f212994a1a.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00073) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fexplaining-in-style?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fexplaining-in-style) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.13369), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.10112), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.12799), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1912.04958), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1710.01711)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fai.googleblog.com\u002F2022\u002F01\u002Fintroducing-stylex-new-approach-for.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fexplaining-in-style.github.io\u002F)\u003C\u002Fli>\u003Cli>[supplementary](https:\u002F\u002Fexplaining-in-style.github.io\u002Fsupmat.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FwLk2eBdXH4M)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fexplaining-in-style\u002Fblob\u002Fmain\u002FExplaining_in_Style_AttFind.ipynb) | 25.08.2021 |\n| VITS | Parallel end-to-end TTS method that generates more natural sounding audio than current two-stage models | \u003Cul>\u003Cli>[Jaehyeon Kim](https:\u002F\u002Fjaywalnut310.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jungil Kong](https:\u002F\u002Fgithub.com\u002Fjik876)\u003C\u002Fli> \u003Cli>[Juhee Son](https:\u002F\u002Fjuheeuu.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjaywalnut310\u002Fvits?style=social)](https:\u002F\u002Fgithub.com\u002Fjaywalnut310\u002Fvits) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.06103)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fjaywalnut310.github.io\u002Fvits-demo\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1CO61pZizDj7en71NQG_aqqKdGaA_SaBf) | 23.08.2021 |\n| Bringing Old Photo Back to Life | Restoring old photos that suffer from severe degradation through a deep learning approach | \u003Cul>\u003Cli>[Ziyu Wan](http:\u002F\u002Fraywzy.com\u002F)\u003C\u002Fli> \u003Cli>[Bo Zhang](https:\u002F\u002Fbo-zhang.me\u002F)\u003C\u002Fli> \u003Cli>[Dongdong Chen](http:\u002F\u002Fwww.dongdongchen.bid\u002F)\u003C\u002Fli> \u003Cli>[Pan Zhang](https:\u002F\u002Fpanzhang0212.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Dong Chen](http:\u002F\u002Fwww.dongchen.pro\u002F)\u003C\u002Fli> \u003Cli>[Jing Liao](https:\u002F\u002Fliaojing.github.io\u002Fhtml\u002F)\u003C\u002Fli> \u003Cli>[Fang Wen](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fpeople\u002Ffangwen\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_5cd53ffe91f1.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00282) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmicrosoft\u002FBringing-Old-Photos-Back-to-Life?style=social)](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FBringing-Old-Photos-Back-to-Life) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.09484)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Freplicate.com\u002Fmicrosoft\u002Fbringing-old-photos-back-to-life)\u003C\u002Fli>\u003Cli>[project](http:\u002F\u002Fraywzy.com\u002FOld_Photo\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FQ5bhszQq9eA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1NEm6AsybIiC5TwTU_4DqDkQO0nFRB-uA) | 13.07.2021 |\n| PTI | Pivotal Tuning Inversion enables employing off-the-shelf latent based semantic editing techniques on real images using StyleGAN | \u003Cul>\u003Cli>[Daniel Roich](https:\u002F\u002Fgithub.com\u002Fdanielroich)\u003C\u002Fli> \u003Cli>[Ron Mokady](https:\u002F\u002Frmokady.github.io\u002F)\u003C\u002Fli> \u003Cli>[Amit Bermano](https:\u002F\u002Fwww.cs.tau.ac.il\u002F~amberman\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_16de55b38de6.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3544777) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdanielroich\u002FPTI?style=social)](https:\u002F\u002Fgithub.com\u002Fdanielroich\u002FPTI) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.05744)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2-ada-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frichzhang\u002FPerceptualSimilarity)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdanielroich\u002FPTI\u002Fblob\u002Fmain\u002Fnotebooks\u002Finference_playground.ipynb) | 01.07.2021 |\n| TediGAN | Framework for multi-modal image generation and manipulation with textual descriptions | \u003Cul>\u003Cli>[Weihao Xia](https:\u002F\u002Fgithub.com\u002Fweihaox)\u003C\u002Fli> \u003Cli>[Yujiu Yang](http:\u002F\u002Fwww.fiesta.tsinghua.edu.cn\u002Fpi\u002F3\u002F24)\u003C\u002Fli> \u003Cli>[Jing-Hao Xue](http:\u002F\u002Fwww.homepages.ucl.ac.uk\u002F~ucakjxu\u002F)\u003C\u002Fli> \u003Cli>[Baoyuan Wu](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fbaoyuanwu2015\u002Fhome)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_dbc9ae17c37e.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00229) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FIIGROUP\u002FTediGAN?style=social)](https:\u002F\u002Fgithub.com\u002FIIGROUP\u002FTediGAN) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.03308), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.08910)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fweihaox\u002FMulti-Modal-CelebA-HQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch\u002F), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffyu\u002Flsun)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FL8Na2f5viAM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](http:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fweihaox\u002FTediGAN\u002Fblob\u002Fmaster\u002Fplayground.ipynb) | 30.06.2021 |\n| SCALE | Modeling Clothed Humans with a Surface Codec of Articulated Local Elements | \u003Cul>\u003Cli>[Qianli Ma](https:\u002F\u002Fqianlim.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shunsuke Saito](https:\u002F\u002Fshunsukesaito.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jinlong Yang](https:\u002F\u002Fis.mpg.de\u002F~jyang)\u003C\u002Fli> \u003Cli>[Siyu Tang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=BUDh_4wAAAAJ)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_380e811aec56.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01582) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fqianlim\u002FSCALE?style=social)](https:\u002F\u002Fgithub.com\u002Fqianlim\u002FSCALE) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.07660)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcape.is.tue.mpg.de\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fkrrish94\u002Fchamferdist), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fshunsukesaito\u002FSCANimate)\u003C\u002Fli>\u003Cli>[poster](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F650\u002FSCALE_poster_CVPR_final_compressed.pdf)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fqianlim.github.io\u002FSCALE.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-EvWqFCUb7U), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fv4rWCxJJzhc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1lp6r-A-s1kBorIvg6rLD4Ja3o6JOvu3G) | 26.06.2021 |\n| CogView | Mastering Text-to-Image Generation via Transformers | \u003Cul>\u003Cli>[Ming Ding](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=Va50YzkAAAAJ)\u003C\u002Fli> \u003Cli>[Zhuoyi Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=tgAt-gEAAAAJ)\u003C\u002Fli> \u003Cli>[Wenyi Hong](https:\u002F\u002Fgithub.com\u002Fwenyihong)\u003C\u002Fli> \u003Cli>[Wendi Zheng](https:\u002F\u002Fgithub.com\u002Fminkowski0125)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Chang Zhou](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=QeSoG3sAAAAJ)\u003C\u002Fli> \u003Cli>[Junyang Lin](https:\u002F\u002Fjustinlin610.github.io\u002F)\u003C\u002Fli> \u003Cli>[Xu Zou](http:\u002F\u002Fxuzou.cn\u002F)\u003C\u002Fli> \u003Cli>[Zhou Shao](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FShao_Zhou4)\u003C\u002Fli> \u003Cli>[Hongxia Yang](https:\u002F\u002Fsites.google.com\u002Fsite\u002Fhystatistics\u002Fhome)\u003C\u002Fli> \u003Cli>[Jie Tang](https:\u002F\u002Fkeg.cs.tsinghua.edu.cn\u002Fjietang\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTHUDM\u002FCogView?style=social)](https:\u002F\u002Fgithub.com\u002FTHUDM\u002FCogView) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2105.13290)\u003C\u002Fli>\u003Cli>[demo](https:\u002F\u002Fthudm.github.io\u002FCogView\u002Findex.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fapex), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FSleepychord\u002Fcogdata)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Ftowardsdatascience.com\u002Fcogview-image-generation-and-language-modelling-at-scale-8d358a0686d2)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2021\u002Fhash\u002Fa4d92e2cd541fca87e4620aba658316d-Abstract.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Fnmxsd8\u002Fr_cogview_mastering_texttoimage_generation_via\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FCw1r8ACIj8U)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Bi2TnSUp2vNiSUhamsNuC4HqkZ2J4WwZ) | 21.06.2021 |\n| GANs N' Roses | Stable, Controllable, Diverse Image to Image Translation | \u003Cul>\u003Cli>[Min Jin Chong](https:\u002F\u002Fmchong6.github.io\u002F)\u003C\u002Fli> \u003Cli>[David Forsyth](http:\u002F\u002Fluthuli.cs.uiuc.edu\u002F~daf\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmchong6\u002FGANsNRoses?style=social)](https:\u002F\u002Fgithub.com\u002Fmchong6\u002FGANsNRoses) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.06561), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2007.06600)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fznxlwm\u002FUGATIT-pytorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVNg0NyCGl_4)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmchong6\u002FGANsNRoses\u002Fblob\u002Fmaster\u002Finference_colab.ipynb) | 19.06.2021 |\n| Rethinking Style Transfer: From Pixels to Parameterized Brushstrokes | A method to stylize images by optimizing parameterized brushstrokes instead of pixels | \u003Cul>\u003Cli>[Dmytro Kotovenko](https:\u002F\u002Fscholar.google.de\u002Fcitations?user=T_U8yxwAAAAJ)\u003C\u002Fli> \u003Cli>[Matthias Wright](https:\u002F\u002Fmatthias-wright.github.io\u002F)\u003C\u002Fli> \u003Cli>[Arthur Heimbrecht](https:\u002F\u002Fgithub.com\u002Farwehei)\u003C\u002Fli> \u003Cli>[Björn Ommer](https:\u002F\u002Fommer-lab.com\u002Fpeople\u002Fommer\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_61bdf3918c27.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01202) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCompVis\u002Fbrushstroke-parameterized-style-transfer?style=social)](https:\u002F\u002Fgithub.com\u002FCompVis\u002Fbrushstroke-parameterized-style-transfer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.17185)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fcompvis.github.io\u002Fbrushstroke-parameterized-style-transfer\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FCompVis\u002Fbrushstroke-parameterized-style-transfer\u002Fblob\u002Ftensorflow_v2\u002Fnotebooks\u002FBrushstrokeStyleTransfer_TF2.ipynb) | 02.06.2021 |\n| Pixel2Style2Pixel | Encoding in Style: A StyleGAN Encoder for Image-to-Image Translation | \u003Cul>\u003Cli>[Elad Richardson](https:\u002F\u002Fgithub.com\u002Feladrich)\u003C\u002Fli> \u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yotam Nitzan](https:\u002F\u002Fyotamnitzan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_4a5992da112b.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00232) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Feladrich\u002Fpixel2style2pixel?style=social)](https:\u002F\u002Fgithub.com\u002Feladrich\u002Fpixel2style2pixel) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.00951)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHuangYG123\u002FCurricularFace)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Feladrich.github.io\u002Fpixel2style2pixel\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbfvSwhqsTgM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Feladrich\u002Fpixel2style2pixel\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 01.06.2021 |\n| Fine-tuning a BERT | We will work through fine-tuning a BERT model using the tensorflow-models PIP package | \u003Cul>\u003Cli>[Chen Chen](https:\u002F\u002Fgithub.com\u002FchenGitHuber)\u003C\u002Fli> \u003Cli>[Claire Yao](https:\u002F\u002Fgithub.com\u002Fclaireyao-fen)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1d8bb9ef5451.png)](https:\u002F\u002Fdoi.org\u002F10.18653\u002Fv1\u002FN19-1423) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.04805)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Ftensorflow.org\u002Fhub)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fmodels\u002Fblob\u002Fmaster\u002Fofficial\u002Fcolab\u002Ffine_tuning_bert.ipynb) | 25.05.2021 |\n| ReStyle | A Residual-Based StyleGAN Encoder via Iterative Refinement | \u003Cul>\u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_4b939d0bae6b.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00664) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyuval-alaluf\u002Frestyle-encoder?style=social)](https:\u002F\u002Fgithub.com\u002Fyuval-alaluf\u002Frestyle-encoder) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.02699), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2008.00951), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.02766)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FTreB1eN\u002FInsightFace_Pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyuval-alaluf.github.io\u002Frestyle-encoder\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyuval-alaluf\u002Frestyle-encoder\u002Fblob\u002Fmaster\u002Fnotebooks\u002Finference_playground.ipynb) | 21.05.2021 |\n| Motion Representations for Articulated Animation | Novel motion representations for animating articulated objects consisting of distinct parts | \u003Cul>\u003Cli>[Aliaksandr Siarohin](https:\u002F\u002Faliaksandrsiarohin.github.io\u002Faliaksandr-siarohin-website\u002F)\u003C\u002Fli> \u003Cli>[Oliver Woodford](https:\u002F\u002Fojwoodford.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jian Ren](https:\u002F\u002Falanspike.github.io\u002F)\u003C\u002Fli> \u003Cli>[Menglei Chai](https:\u002F\u002Fmlchai.com\u002F)\u003C\u002Fli> \u003Cli>[Sergey Tulyakov](http:\u002F\u002Fwww.stulyakov.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_852569f5d4b8.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01344) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsnap-research\u002Farticulated-animation?style=social)](https:\u002F\u002Fgithub.com\u002Fsnap-research\u002Farticulated-animation) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.11280)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fsnap-research.github.io\u002Farticulated-animation\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=gpBYN8t8_yY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAliaksandrSiarohin\u002Farticulated-animation\u002Fblob\u002Fmaster\u002Fdemo.ipynb) | 29.04.2021 |\n| SAM | Age Transformation Using a Style-Based Regression Model | \u003Cul>\u003Cli>[Yuval Alaluf](https:\u002F\u002Fyuval-alaluf.github.io\u002F)\u003C\u002Fli> \u003Cli>[Or Patashnik](https:\u002F\u002Forpatashnik.github.io\u002F)\u003C\u002Fli> \u003Cli>[Daniel Cohen-Or](https:\u002F\u002Fdanielcohenor.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0239aadbb33a.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3450626.3459805) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyuval-alaluf\u002FSAM?style=social)](https:\u002F\u002Fgithub.com\u002Fyuval-alaluf\u002FSAM) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.02754)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Feladrich\u002Fpixel2style2pixel), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fyuval-alaluf.github.io\u002FSAM\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FX_pYC_LtBFw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](http:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyuval-alaluf\u002FSAM\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fanimation_inference_playground.ipynb) | 26.04.2021 |\n| Geometry-Free View Synthesis | Is a geometric model required to synthesize novel views from a single image? | \u003Cul>\u003Cli>[Robin Rombach](https:\u002F\u002Fgithub.com\u002Frromb)\u003C\u002Fli> \u003Cli>[Patrick Esser](https:\u002F\u002Fgithub.com\u002Fpesser)\u003C\u002Fli> \u003Cli>[Björn Ommer](https:\u002F\u002Fommer-lab.com\u002Fpeople\u002Fommer\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_9f50cfdb8dc3.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.01409) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FCompVis\u002Fgeometry-free-view-synthesis?style=social)](https:\u002F\u002Fgithub.com\u002FCompVis\u002Fgeometry-free-view-synthesis) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2104.07652)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fgoogle.github.io\u002Frealestate10k\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcolmap\u002Fcolmap)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FCompVis\u002Fgeometry-free-view-synthesis\u002Fblob\u002Fmaster\u002Fscripts\u002Fbraindance.ipynb) | 22.04.2021 |\n| NeRViS | An algorithm for full-frame video stabilization by first estimating dense warp fields | \u003Cul>\u003Cli>[Yu-Lun Liu](http:\u002F\u002Fwww.cmlab.csie.ntu.edu.tw\u002F~yulunliu\u002F)\u003C\u002Fli> \u003Cli>[Wei-Sheng Lai](https:\u002F\u002Fwww.wslai.net\u002F)\u003C\u002Fli> \u003Cli>[Ming-Hsuan Yang](https:\u002F\u002Ffaculty.ucmerced.edu\u002Fmhyang\u002F)\u003C\u002Fli> \u003Cli>[Yung-Yu Chuang](https:\u002F\u002Fwww.csie.ntu.edu.tw\u002F~cyy\u002F)\u003C\u002Fli> \u003Cli>[Jia-Bin Huang](https:\u002F\u002Fjbhuang0604.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fd196cdcdbc9.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV48922.2021.00230) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Falex04072000\u002FNeRViS?style=social)](https:\u002F\u002Fgithub.com\u002Falex04072000\u002FNeRViS) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.06205)\u003C\u002Fli>\u003Cli>[data](http:\u002F\u002Fliushuaicheng.org\u002FSIGGRAPH2013\u002Fdatabase.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcxjyxxme\u002Fdeep-online-video-stabilization), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fjinsc37\u002FDIFRINT)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Falex04072000.github.io\u002FNeRViS\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FKO3sULs4hso)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1l-fUzyM38KJMZyKMBWw_vu7ZUyDwgdYH) | 11.04.2021 |\n| NeX | View synthesis based on enhancements of multiplane image that can reproduce NeXt-level view-dependent effects in real time | \u003Cul>\u003Cli>[Suttisak Wizadwongsa](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsuttisak-wizadwongsa-763a931a5\u002F)\u003C\u002Fli> \u003Cli>[Pakkapon Phongthawee](http:\u002F\u002Fpureexe.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jiraphon Yenphraphai](https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fjiraphon-yenphraphai-990ba6175\u002F)\u003C\u002Fli> \u003Cli>[Supasorn Suwajanakorn](https:\u002F\u002Fwww.supasorn.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_c42449f21921.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00843) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fnex-mpi\u002Fnex-code?style=social)](https:\u002F\u002Fgithub.com\u002Fnex-mpi\u002Fnex-code) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.05606)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fvistec-my.sharepoint.com\u002Fpersonal\u002Fpakkapon_p_s19_vistec_ac_th\u002F_layouts\u002F15\u002Fonedrive.aspx?id=%2Fpersonal%2Fpakkapon%5Fp%5Fs19%5Fvistec%5Fac%5Fth%2FDocuments%2Fpublic%2FVLL%2FNeX%2Fshiny%5Fdatasets&originalPath=aHR0cHM6Ly92aXN0ZWMtbXkuc2hhcmVwb2ludC5jb20vOmY6L2cvcGVyc29uYWwvcGFra2Fwb25fcF9zMTlfdmlzdGVjX2FjX3RoL0VuSVVoc1JWSk9kTnNaXzRzbWRoeWUwQjh6MFZseHFPUjM1SVIzYnAwdUd1cFE%5FcnRpbWU9WXRVQTQtQTcyVWc), [data](https:\u002F\u002Fvistec-my.sharepoint.com\u002Fpersonal\u002Fpakkapon_p_s19_vistec_ac_th\u002F_layouts\u002F15\u002Fonedrive.aspx?originalPath=aHR0cHM6Ly92aXN0ZWMtbXkuc2hhcmVwb2ludC5jb20vOmY6L2cvcGVyc29uYWwvcGFra2Fwb25fcF9zMTlfdmlzdGVjX2FjX3RoL0VyalBSUkw5Sm5GSXA4TU42ZDFqRXVvQjNYVm94SmtmZlBqZm9QeWhIa2owZGc%5FcnRpbWU9bC0yYWctRTcyVWc&id=%2Fpersonal%2Fpakkapon%5Fp%5Fs19%5Fvistec%5Fac%5Fth%2FDocuments%2Fpublic%2FVLL%2FNeX%2Fmodified%5Fdataset)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FFyusion\u002FLLFF)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fnex-mpi.github.io\u002F)\u003C\u002Fli>\u003Cli>[vistec](https:\u002F\u002Fvistec.ist\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=HyfkF7Z-ddA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1hXVvYdAwLA0EFg2zrafJUE0bFgB_F7PU) | 25.03.2021 |\n| Score SDE | Score-Based Generative Modeling through Stochastic Differential Equations | \u003Cul>\u003Cli>[Yang Song](https:\u002F\u002Fyang-song.net\u002F)\u003C\u002Fli> \u003Cli>[Jascha Sohl-Dickstein](http:\u002F\u002Fwww.sohldickstein.com\u002F)\u003C\u002Fli> \u003Cli>[Diederik Kingma](http:\u002F\u002Fdpkingma.com\u002F)\u003C\u002Fli> \u003Cli>[Abhishek Kumar](https:\u002F\u002Fabhishek.umiacs.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Stefano Ermon](https:\u002F\u002Fcs.stanford.edu\u002F~ermon\u002F)\u003C\u002Fli> \u003Cli>[Ben Poole](https:\u002F\u002Fcs.stanford.edu\u002F~poole\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyang-song\u002Fscore_sde?style=social)](https:\u002F\u002Fgithub.com\u002Fyang-song\u002Fscore_sde) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.13456), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.05600), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.09011), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.11239)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fyang-song\u002Fscore_sde_pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fml_collections)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FL9ZegT87QK8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fyang-song\u002Fscore_sde\u002Fblob\u002Fmain\u002FScore_SDE_demo.ipynb) | 18.03.2021 |\n| Talking Head Anime from a Single Image | The network takes as input an image of an anime character's face and a desired pose, and it outputs another image of the same character in the given pose | [Pramook Khungurn](https:\u002F\u002Fpkhungurn.github.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fpkhungurn\u002Ftalking-head-anime-demo?style=social)](https:\u002F\u002Fgithub.com\u002Fpkhungurn\u002Ftalking-head-anime-demo) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flincolnhard\u002Fhead-pose-estimation)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fpkhungurn.github.io\u002Ftalking-head-anime\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FVirtual_YouTuber), [\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FMikuMikuDance)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FkMQCERkTdO0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FT1Gp-RxFZwU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FFioRJ6x_RbI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fpkhungurn\u002Ftalking-head-anime-demo\u002Fblob\u002Fmaster\u002Ftha_colab.ipynb) | 23.02.2021 |\n| NFNet | An adaptive gradient clipping technique, a significantly improved class of Normalizer-Free ResNets | \u003Cul>\u003Cli>[Andrew Brock](https:\u002F\u002Fgithub.com\u002Fajbrock)\u003C\u002Fli> \u003Cli>[Soham De](https:\u002F\u002Fsohamde.github.io\u002F)\u003C\u002Fli> \u003Cli>[Samuel L. Smith](https:\u002F\u002Fscholar.google.co.uk\u002Fcitations?user=fyEqU5oAAAAJ)\u003C\u002Fli> \u003Cli>[Karen Simonyan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=L7lMQkQAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepmind\u002Fdeepmind-research?style=social)](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fdeepmind-research\u002Ftree\u002Fmaster\u002Fnfnets) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.06171), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2101.08692)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepmind\u002Fjaxline)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FrNkHjZtH0RQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Flive\u002Fqyy2WhRRSI4?feature=share)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdeepmind\u002Fdeepmind-research\u002Fblob\u002Fmaster\u002Fnfnets\u002Fnfnet_demo_colab.ipynb) | 17.02.2021 |\n| RITM | Simple feedforward model for click-based interactive segmentation that employs the segmentation masks from previous steps | \u003Cul>\u003Cli>[Konstantin Sofiiuk](https:\u002F\u002Fgithub.com\u002Fksofiyuk)\u003C\u002Fli> \u003Cli>[Ilia Petrov](https:\u002F\u002Fvirtualhumans.mpi-inf.mpg.de\u002Fpeople\u002FPetrov.html)\u003C\u002Fli> \u003Cli>[Anton Konushin](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=ZT_k-wMAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_81755896538e.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICIP46576.2022.9897365) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fsupervisely-ecosystem\u002Fritm-interactive-segmentation?style=social)](https:\u002F\u002Fgithub.com\u002Fsupervisely-ecosystem\u002Fritm-interactive-segmentation) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2102.06583)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FHRNet\u002FHRNet-Image-Classification)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Finteractive-segmentation-on-grabcut?p=reviving-iterative-training-with-mask), [\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Finteractive-segmentation-on-berkeley?p=reviving-iterative-training-with-mask)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fsupervisely-ecosystem\u002Fritm_interactive_segmentation\u002Fblob\u002Fmaster\u002Fnotebooks\u002Fcolab_test_any_model.ipynb) | 13.02.2021 |\n| CLIP | A neural network which efficiently learns visual concepts from natural language supervision | \u003Cul>\u003Cli>[Jong Wook Kim](https:\u002F\u002Fjongwook.kim\u002F)\u003C\u002Fli> \u003Cli>[Alec Radford](http:\u002F\u002Fnewmu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ilya Sutskever](http:\u002F\u002Fwww.cs.utoronto.ca\u002F~ilya\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fopenai\u002FCLIP?style=social)](https:\u002F\u002Fgithub.com\u002Fopenai\u002FCLIP) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2103.00020)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fwww.cs.toronto.edu\u002F~kriz\u002Fcifar.html)\u003C\u002Fli>\u003Cli>[paper](https:\u002F\u002Fcdn.openai.com\u002Fpapers\u002FLearning_Transferable_Visual_Models_From_Natural_Language_Supervision.pdf)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fopenai.com\u002Fblog\u002Fclip\u002F)\u003C\u002Fli>\u003Cli>[slides](https:\u002F\u002Ficml.cc\u002Fmedia\u002Ficml-2021\u002FSlides\u002F9193.pdf)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fopenai\u002Fclip\u002Fblob\u002Fmaster\u002FInteracting_with_CLIP.ipynb) | 29.01.2021 |\n| Adversarial Patch | A method to create universal, robust, targeted adversarial image patches in the real world | [Tom Brown](https:\u002F\u002Fgithub.com\u002Fnottombrown) | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1712.09665)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fcleverhans-lab\u002Fcleverhans\u002Fblob\u002Fmaster\u002Fexamples\u002Fadversarial_patch\u002FAdversarialPatch.ipynb) | 27.01.2021 |\n| MSG-Net | Multi-style Generative Network with a novel Inspiration Layer, which retains the functionality of optimization-based approaches and has the fast speed of feed-forward networks | \u003Cul>\u003Cli>[Hang Zhang](https:\u002F\u002Fhangzhang.org\u002F)\u003C\u002Fli> \u003Cli>[Kristin Dana](https:\u002F\u002Fwww.ece.rutgers.edu\u002F~kdana\u002Fdana.html)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_81755896538e.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-11018-5_32) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1703.06953)\u003C\u002Fli>\u003Cli>[project](http:\u002F\u002Fcomputervisionrutgers.github.io\u002FMSG-Net\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=oy6pWNWBt4Y)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fzhanghang1989\u002FPyTorch-Multi-Style-Transfer\u002Fblob\u002Fmaster\u002Fmsgnet.ipynb) | 25.01.2021 |\n| Neural Style Transfer | Implementation of Neural Style Transfer in Keras 2.0+ | [Somshubra Majumdar](http:\u002F\u002Ftitu1994.github.io\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_73994007fd6b.png)](https:\u002F\u002Fdoi.org\u002F10.1167\u002F16.12.326) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftitu1994\u002FNeural-Style-Transfer?style=social)](https:\u002F\u002Fgithub.com\u002Ftitu1994\u002FNeural-Style-Transfer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F1508.06576), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F1605.04603), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1606.05897)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftitu1994\u002FNeural-Style-Transfer\u002Fblob\u002Fmaster\u002FNeuralStyleTransfer.ipynb) | 22.01.2021 |\n| SkyAR | A vision-based method for video sky replacement and harmonization, which can automatically generate realistic and dramatic sky backgrounds in videos with controllable styles | [Zhengxia Zou](http:\u002F\u002Fwww-personal.umich.edu\u002F~zzhengxi\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_612548bd9a37.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTIP.2022.3192717) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjiupinjia\u002FSkyAR?style=social)](https:\u002F\u002Fgithub.com\u002Fjiupinjia\u002FSkyAR) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.11800)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fjiupinjia.github.io\u002Fskyar\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=zal9Ues0aOQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fjiupinjia\u002FSkyAR\u002Fblob\u002Fmaster\u002Fcolab_demo.ipynb) | 18.01.2021 |\n| MusicXML Documentation | The goal of this notebook is to explore one of the magenta libraries for music | \u003Cul>\u003Cli>[Prakruti Joshi](https:\u002F\u002Fgithub.com\u002Fprakruti-joshi)\u003C\u002Fli> \u003Cli>[Falak Shah](https:\u002F\u002Ffalaktheoptimist.github.io\u002F)\u003C\u002Fli> \u003Cli>[Twisha Naik](https:\u002F\u002Fgithub.com\u002Ftwisha96)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>[magenta](https:\u002F\u002Fmagenta.tensorflow.org\u002F)\u003C\u002Fli>\u003Cli>[music theory](http:\u002F\u002Fmusictheoryblog.blogspot.com\u002F2008\u002F02\u002Flearn-music-theory.html)\u003C\u002Fli>\u003Cli>[musicXML](https:\u002F\u002Fwww.musicxml.com\u002Ffor-developers\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmagenta\u002Fmagenta-demos\u002Fblob\u002Fmaster\u002Fcolab-notebooks\u002FMusicXML_Document_Structure_Documentation.ipynb) | 08.01.2021 |\n| SVG VAE | A colab demo for the SVG VAE model | [Raphael Gontijo Lopes](https:\u002F\u002Fraphagl.com\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_3ad9d536e095.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICCV.2019.00802) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.02632)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fmagenta.tensorflow.org\u002Fsvg-vae)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmagenta\u002Fmagenta-demos\u002Fblob\u002Fmaster\u002Fcolab-notebooks\u002Fvae_svg_decoding.ipynb) | 08.01.2021 |\n| Neural Magic Eye | Learning to See and Understand the Scene Behind an Autostereogram | \u003Cul>\u003Cli>[Zhengxia Zou](http:\u002F\u002Fwww-personal.umich.edu\u002F~zzhengxi\u002F)\u003C\u002Fli> \u003Cli>[Tianyang Shi](https:\u002F\u002Fwww.shitianyang.tech\u002F)\u003C\u002Fli> \u003Cli>[Yi Yuan](https:\u002F\u002Fyiyuan1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhenwei Shi](http:\u002F\u002Flevir.buaa.edu.cn\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjiupinjia\u002Fneural-magic-eye?style=social)](https:\u002F\u002Fgithub.com\u002Fjiupinjia\u002Fneural-magic-eye) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.15692)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fjiupinjia.github.io\u002Fneuralmagiceye\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=Fkh7DEblqJ8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1f59dFLJ748i2TleE54RkbUZSMo9Hyx7l) | 01.01.2021 |\n| FGVC | Method first extracts and completes motion edges, and then uses them to guide piecewise-smooth flow completion with sharp edges | \u003Cul>\u003Cli>[Chen Gao](http:\u002F\u002Fchengao.vision\u002F)\u003C\u002Fli> \u003Cli>[Ayush Saraf](https:\u002F\u002Fgithub.com\u002Fayush29feb)\u003C\u002Fli> \u003Cli>[Johannes Kopf](https:\u002F\u002Fjohanneskopf.de\u002F)\u003C\u002Fli> \u003Cli>[Jia-Bin Huang](https:\u002F\u002Fjbhuang0604.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2af4fe162fb6.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-58610-2_42) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fvt-vl-lab\u002FFGVC?style=social)](https:\u002F\u002Fgithub.com\u002Fvt-vl-lab\u002FFGVC) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2009.01835)\u003C\u002Fli>\u003Cli>[project](http:\u002F\u002Fchengao.vision\u002FFGVC\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=CHHVPxHT7rc)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1pb6FjWdwq_q445rG2NP0dubw7LKNUkqc) | 30.12.2020 |\n| VIBE | Video Inference for Body Pose and Shape Estimation, which makes use of an existing large-scale motion capture dataset together with unpaired, in-the-wild, 2D keypoint annotations | \u003Cul>\u003Cli>[Muhammed Kocabas](https:\u002F\u002Fps.is.mpg.de\u002Fperson\u002Fmkocabas)\u003C\u002Fli> \u003Cli>[Nikos Athanasiou](https:\u002F\u002Fgithub.com\u002Fathn-nik)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002Fperson\u002Fblack)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ed0b30996d1b.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00530) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmkocabas\u002FVIBE?style=social)](https:\u002F\u002Fgithub.com\u002Fmkocabas\u002FVIBE) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1912.05656)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fcarlosedubarreto\u002Fvibe_win_install), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fakanazawa\u002Fhuman_dynamics), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMandyMo\u002Fpytorch_HMR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fsoulslicer\u002FSTAF\u002Ftree\u002Fstaf)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002F3d-human-pose-estimation-on-3dpw?p=vibe-video-inference-for-human-body-pose-and)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3qhs5IRJ1LI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fw1biKeiQThY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FrIr-nX63dUA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfW0sIZfQcIs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8Qt0wA16kTo), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fxyo5gl5GLEI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FXNzgUhxKC38), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FhErK0MamTY4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGfmm8uMfMq0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1dFfwxZ52MN86FA6uFNypMEdFShd2euQA) | 23.12.2020 |\n| SeFa | A closed-form approach for unsupervised latent semantic factorization in GANs | \u003Cul>\u003Cli>[Yujun Shen](https:\u002F\u002Fshenyujun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bolei Zhou](https:\u002F\u002Fboleizhou.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a10ddf728c4c.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.00158) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgenforce\u002Fsefa?style=social)](https:\u002F\u002Fgithub.com\u002Fgenforce\u002Fsefa) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2007.06600)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgenforce.github.io\u002Fsefa\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=OFHW2WbXXIQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgenforce\u002Fsefa\u002Fblob\u002Fmaster\u002Fdocs\u002FSeFa.ipynb) | 06.12.2020 |\n| Stylized Neural Painting | An image-to-painting translation method that generates vivid and realistic painting artworks with controllable styles | \u003Cul>\u003Cli>[Zhengxia Zou](http:\u002F\u002Fwww-personal.umich.edu\u002F~zzhengxi\u002F)\u003C\u002Fli> \u003Cli>[Tianyang Shi](https:\u002F\u002Fwww.shitianyang.tech\u002F)\u003C\u002Fli> \u003Cli>[Yi Yuan](https:\u002F\u002Fyiyuan1991.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zhenwei Shi](http:\u002F\u002Flevir.buaa.edu.cn\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_52b6cddbd876.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR46437.2021.01543) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fjiupinjia\u002Fstylized-neural-painting?style=social)](https:\u002F\u002Fgithub.com\u002Fjiupinjia\u002Fstylized-neural-painting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.08114)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fjiupinjia.github.io\u002Fneuralpainter\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=oerb-nwrXhk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1ch_41GtcQNQT1NLOA21vQJ_rQOjjv9D8) | 01.12.2020 |\n| BiT | Big Transfer: General Visual Representation Learning | \u003Cul>\u003Cli>[Alexander Kolesnikov](https:\u002F\u002Fgithub.com\u002Fakolesnikoff)\u003C\u002Fli> \u003Cli>[Lucas Beyer](http:\u002F\u002Flucasb.eyer.be)\u003C\u002Fli> \u003Cli>[Xiaohua Zhai](https:\u002F\u002Fgithub.com\u002Fxiaohuazhai)\u003C\u002Fli> \u003Cli>[Joan Puigcerver](https:\u002F\u002Fwww.jpuigcerver.net\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Jessica Yung](https:\u002F\u002Fgithub.com\u002Fjessicayung)\u003C\u002Fli> \u003Cli>[Sylvain Gelly](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=m7LvuTkAAAAJ)\u003C\u002Fli> \u003Cli>[Neil Houlsby](https:\u002F\u002Fneilhoulsby.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_37c60b5e4977.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-58558-7_29) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fbig_transfer?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fbig_transfer) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1912.11370), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2106.05237)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fgoogle\u002Fbit-50)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Freview-big-transfer-bit-general-visual-representation-learning-cb4bf8ed9732)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fk1GOF2jmX7c), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F0iTgt5-SOsU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FX5Rhm__OxvA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Fbig_transfer\u002Fblob\u002Fmaster\u002Fcolabs\u002Fbig_transfer_tf2.ipynb) | 12.11.2020 |\n| LaSAFT | Latent Source Attentive Frequency Transformation for Conditioned Source Separation | [Woosung Choi](https:\u002F\u002Fws-choi.github.io\u002F) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_4563a7746cdb.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICASSP39728.2021.9413896) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fws-choi\u002FConditioned-Source-Separation-LaSAFT?style=social)](https:\u002F\u002Fgithub.com\u002Fws-choi\u002FConditioned-Source-Separation-LaSAFT) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.11631)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fsigsep.github.io\u002Fdatasets\u002Fmusdb.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Flasaft.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fws-choi\u002FConditioned-Source-Separation-LaSAFT\u002Fblob\u002Fmaster\u002Fcolab_demo\u002FLaSAFT_with_GPoCM_Stella_Jang_Example.ipynb) | 01.11.2020 |\n| Lifespan Age Transformation Synthesis | Multi-domain image-to-image generative adversarial network architecture, whose learned latent space models a continuous bi-directional aging process | \u003Cul>\u003Cli>[Roy Or-El](https:\u002F\u002Fhomes.cs.washington.edu\u002F~royorel\u002F)\u003C\u002Fli> \u003Cli>[Soumyadip Sengupta](https:\u002F\u002Fhomes.cs.washington.edu\u002F~soumya91\u002F)\u003C\u002Fli> \u003Cli>[Ohad Fried](https:\u002F\u002Fwww.ohadf.com\u002F)\u003C\u002Fli> \u003Cli>[Eli Shechtman](https:\u002F\u002Fresearch.adobe.com\u002Fperson\u002Feli-shechtman\u002F)\u003C\u002Fli> \u003Cli>[Ira Kemelmacher-Shlizerman](https:\u002F\u002Fwww.irakemelmacher.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_083d7e30841c.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-58539-6_44) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Froyorel\u002FLifespan_Age_Transformation_Synthesis?style=social)](https:\u002F\u002Fgithub.com\u002Froyorel\u002FLifespan_Age_Transformation_Synthesis) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.09764)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Froyorel\u002FFFHQ-Aging-Dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fpix2pixHD), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstyle-based-gan-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgrail.cs.washington.edu\u002Fprojects\u002Flifespan_age_transformation_synthesis\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_jTFcjN2hBk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F9fulnt2_q_Y)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Froyorel\u002FLifespan_Age_Transformation_Synthesis\u002Fblob\u002Fmaster\u002FLATS_demo.ipynb) | 31.10.2020 |\n| IDInvert | In-domain GAN inversion approach, which not only faithfully reconstructs the input image but also ensures the inverted code to be semantically meaningful for editing | \u003Cul>\u003Cli>[Jiapeng Zhu](https:\u002F\u002Fgithub.com\u002Fzhujiapeng)\u003C\u002Fli> \u003Cli>[Yujun Shen](https:\u002F\u002Fshenyujun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Deli Zhao](https:\u002F\u002Fzhaodeli.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bolei Zhou](https:\u002F\u002Fboleizhou.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1ba54266a086.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-58520-4_35) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgenforce\u002Fidinvert?style=social)](https:\u002F\u002Fgithub.com\u002Fgenforce\u002Fidinvert) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.00049)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgenforce\u002Fidinvert_pytorch), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffyu\u002Flsun), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ffyu\u002Flsun), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fai-innovation\u002Fin-domain-gan-inversion-for-anime-character-f6341d72e835)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgenforce.github.io\u002Fidinvert\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F2qMw8sOsNg0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3v6NHrhuyFY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTVSJO9uNq7g)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgenforce\u002Fidinvert_pytorch\u002Fblob\u002Fmaster\u002Fdocs\u002FIdinvert.ipynb) | 22.10.2020 |\n| HiGAN | Semantic Hierarchy Emerges in Deep Generative Representations for Scene Synthesis | \u003Cul>\u003Cli>[Ceyuan Yang](https:\u002F\u002Fceyuan.me\u002F)\u003C\u002Fli> \u003Cli>[Yujun Shen](https:\u002F\u002Fshenyujun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bolei Zhou](https:\u002F\u002Fboleizhou.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_e9598715f8e3.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs11263-020-01429-5) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgenforce\u002Fhigan?style=social)](https:\u002F\u002Fgithub.com\u002Fgenforce\u002Fhigan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1911.09267), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1412.6856), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.10112)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgenforce.github.io\u002Fhigan\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=X5yWu2Jwjpg)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgenforce\u002Fhigan\u002Fblob\u002Fmaster\u002Fdocs\u002FHiGAN_Bedroom.ipynb) | 14.10.2020 |\n| InterFaceGAN | Interpreting the Latent Space of GANs for Semantic Face Editing | \u003Cul>\u003Cli>[Yujun Shen](https:\u002F\u002Fshenyujun.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jinjin Gu](https:\u002F\u002Fwww.jasongt.com\u002F)\u003C\u002Fli> \u003Cli>[Xiaoou Tang](https:\u002F\u002Fwww.ie.cuhk.edu.hk\u002Fpeople\u002Fxotang.shtml)\u003C\u002Fli> \u003Cli>[Bolei Zhou](https:\u002F\u002Fboleizhou.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_27a1ab666155.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00926) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgenforce\u002Finterfacegan?style=social)](https:\u002F\u002Fgithub.com\u002Fgenforce\u002Finterfacegan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.10786), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.09635), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1710.10196)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Ftkarras\u002Fprogressive_growing_of_gans), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fgenforce.github.io\u002Finterfacegan\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=uoftpl3Bj6w)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgenforce\u002Finterfacegan\u002Fblob\u002Fmaster\u002Fdocs\u002FInterFaceGAN.ipynb) | 13.10.2020 |\n| Instance-aware Image Colorization | Novel deep learning framework to achieve instance-aware colorization | [Jheng-Wei Su](https:\u002F\u002Fgithub.com\u002Fericsujw) | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_0d2de5cb9566.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00799) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fericsujw\u002FInstColorization?style=social)](https:\u002F\u002Fgithub.com\u002Fericsujw\u002FInstColorization) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.10825)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fericsujw.github.io\u002FInstColorization\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=Zj1N4uE1ehk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fericsujw\u002FInstColorization\u002Fblob\u002Fmaster\u002FInstColorization.ipynb) | 30.08.2020 |\n| MnasNet | Automated mobile neural architecture search approach, which explicitly incorporate model latency into the main objective so that the search can identify a model that achieves a good trade-off between accuracy and latency | \u003Cul>\u003Cli>[Mingxing Tan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=6POeyBoAAAAJ)\u003C\u002Fli> \u003Cli>[Bo Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=HRDIoP0AAAAJ)\u003C\u002Fli> \u003Cli>[Ruoming Pang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=1fsmwB8AAAAJ)\u003C\u002Fli> \u003Cli>[Vijay Vasudevan](https:\u002F\u002Fvijay.vasu.org)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Mark Sandler](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=IcPc-OUAAAAJ)\u003C\u002Fli> \u003Cli>[Andrew Howard](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=_9l8vD8AAAAJ)\u003C\u002Fli> \u003Cli>[Quoc Le](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=vfT6-XIAAAJ)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_caae2795076f.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR.2019.00293) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Ftpu?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Ftpu\u002Ftree\u002Fmaster\u002Fmodels\u002Fofficial\u002Fmnasnet) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1807.11626)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAnjieCheng\u002FMnasNet-PyTorch)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fanalytics-vidhya\u002Fan-overview-on-mnasnet-platform-aware-neural-architecture-search-for-mobile-8a681d17a80c)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fdocs.pytorch.org\u002Fvision\u002Fmain\u002Fmodels\u002Fmnasnet.html)\u003C\u002Fli>\u003Cli>[tutorial](https:\u002F\u002Fcloud.google.com\u002Ftpu\u002Fdocs\u002Ftutorials\u002Fmnasnet)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F4uDZxefPd-I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F9hHKgPg4Wy0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdOwS37yZSew), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRkCa8OHSF9w), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F9kqfz0qPSW8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Ftpu\u002Fblob\u002Fmaster\u002Fmodels\u002Fofficial\u002Fmnasnet\u002Fmnasnet_example.ipynb) | 27.08.2020 |\n| MoCo | Momentum Contrast for unsupervised visual representation learning | \u003Cul>\u003Cli>[Kaiming He](https:\u002F\u002Fkaiminghe.github.io\u002F)\u003C\u002Fli> \u003Cli>[Haoqi Fan](https:\u002F\u002Fhaoqifan.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yuxin Wu](https:\u002F\u002Fppwwyyxx.com\u002F)\u003C\u002Fli> \u003Cli>[Saining Xie](http:\u002F\u002Fsainingxie.com\u002F)\u003C\u002Fli> \u003Cli>[Ross Girshick](https:\u002F\u002Fwww.rossgirshick.info\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_ff14d4f09568.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00975) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ffacebookresearch\u002Fmoco?style=social)](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fmoco) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1911.05722), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.04297), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.02677)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fppwwyyxx\u002Fmoco.tensorflow)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FLvHwBQF14zs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F4VVGtYPM8JE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fo5Qh61dLDf0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ffacebookresearch\u002Fmoco\u002Fblob\u002Fcolab-notebook\u002Fcolab\u002Fmoco_cifar10_demo.ipynb) | 20.08.2020 |\n| CAPE | Learning to Dress 3D People in Generative Clothing | \u003Cul>\u003Cli>[Qianli Ma](https:\u002F\u002Fqianlim.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jinlong Yang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=HGt39SUAAAAJ)\u003C\u002Fli> \u003Cli>[Anurag Ranjan](https:\u002F\u002Fanuragranj.github.io\u002F)\u003C\u002Fli> \u003Cli>[Sergi Pujades](https:\u002F\u002Fgithub.com\u002Fpujades)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Gerard Pons-Moll](https:\u002F\u002Fvirtualhumans.mpi-inf.mpg.de\u002F)\u003C\u002Fli> \u003Cli>[Siyu Tang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=BUDh_4wAAAAJ)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002F~black)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_fc622cc59a15.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00650) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fqianlim\u002FCAPE?style=social)](https:\u002F\u002Fgithub.com\u002Fqianlim\u002FCAPE) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.13615), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1807.10267), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.02658)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fcape.is.tue.mpg.de\u002Fdataset)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMPI-IS\u002Fmesh), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplx), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fanuragranj\u002Fcoma)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@mahyarfardinfar\u002Flearning-to-dress-3d-people-in-generative-clothing-486eb90136ff)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fcape.is.tue.mpg.de\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fe4W-hPFNwDE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FNOEA-Rtq6vM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1DCNo2OyyTNi1xDG-7j32FZQ9sBA6i9Ys) | 05.08.2020 |\n| Rewriting a Deep Generative Model | We ask if a deep network can be reprogrammed to follow different rules, by enabling a user to directly change the weights, instead of training with a data set | \u003Cul>\u003Cli>[David Bau](https:\u002F\u002Fpeople.csail.mit.edu\u002Fdavidbau\u002Fhome\u002F)\u003C\u002Fli> \u003Cli>[Steven Liu](http:\u002F\u002Fpeople.csail.mit.edu\u002Fstevenliu\u002F)\u003C\u002Fli> \u003Cli>[Tongzhou Wang](https:\u002F\u002Fssnl.github.io\u002F)\u003C\u002Fli> \u003Cli>[Jun-Yan Zhu](https:\u002F\u002Fwww.cs.cmu.edu\u002F~junyanz\u002F)\u003C\u002Fli> \u003Cli>[Antonio Torralba](https:\u002F\u002Fgroups.csail.mit.edu\u002Fvision\u002Ftorralbalab\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_b5b78a288583.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002F978-3-030-58452-8_21) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdavidbau\u002Frewriting?style=social)](https:\u002F\u002Fgithub.com\u002Fdavidbau\u002Frewriting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2007.15646), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1912.04958)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan2), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frosinality\u002Fstylegan2-pytorch)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Frewriting.csail.mit.edu\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=i2_-zNqtEPk), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Frewriting.csail.mit.edu\u002Fvideo\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fdavidbau\u002Frewriting\u002Fblob\u002Fmaster\u002Fnotebooks\u002Frewriting-interface.ipynb) | 01.08.2020 |\n| SIREN | Implicit Neural Representations with Periodic Activation Functions | \u003Cul>\u003Cli>[Vincent Sitzmann](https:\u002F\u002Fvsitzmann.github.io\u002F)\u003C\u002Fli> \u003Cli>[Julien Martel](http:\u002F\u002Fweb.stanford.edu\u002F~jnmartel\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fvsitzmann\u002Fsiren?style=social)](https:\u002F\u002Fgithub.com\u002Fvsitzmann\u002Fsiren) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2006.09661)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fdrive.google.com\u002Fdrive\u002Ffolders\u002F1_iq__37-hw7FJOEUK1tX7mdp8SKB368K)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper\u002F2020\u002Fhash\u002F53c04118df112c13a8c34b38343b9c10-Abstract.html)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fvsitzmann.github.io\u002Fsiren\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=Q2fLWGBeaiI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fvsitzmann\u002Fsiren\u002Fblob\u002Fmaster\u002Fexplore_siren.ipynb) | 25.06.2020 |\n| 3D Photo Inpainting | Method for converting a single RGB-D input image into a 3D photo, i.e., a multi-layer representation for novel view synthesis that contains hallucinated color and depth structures in regions occluded in the original view | \u003Cul>\u003Cli>[Meng-Li Shih](https:\u002F\u002Fshihmengli.github.io\u002F)\u003C\u002Fli> \u003Cli>[Shih-Yang Su](https:\u002F\u002Flemonatsu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Johannes Kopf](https:\u002F\u002Fjohanneskopf.de\u002F)\u003C\u002Fli> \u003Cli>[Jia-Bin Huang](https:\u002F\u002Fjbhuang0604.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_77288fc3bc90.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR42600.2020.00805) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fvt-vl-lab\u002F3d-photo-inpainting?style=social)](https:\u002F\u002Fgithub.com\u002Fvt-vl-lab\u002F3d-photo-inpainting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.04727)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fshihmengli.github.io\u002F3D-Photo-Inpainting\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1706ToQrkIZshRSJSHvZ1RuCiM__YX3Bz) | 04.05.2020 |\n| Audio-driven Talking Face Video Generation with Learning-based Personalized Head Pose | Deep neural network model that takes an audio signal A of a source person and a very short video V of a target person as input, and outputs a synthesized high-quality talking face video with personalized head pose (making use of the visual information in V), expression and lip synchronization (by considering both A and V) | \u003Cul>\u003Cli>[Ran Yi](https:\u002F\u002Fyiranran.github.io\u002F)\u003C\u002Fli> \u003Cli>[Zipeng Ye](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=faIXk9EAAAAJ)\u003C\u002Fli> \u003Cli>[Juyong Zhang](https:\u002F\u002Fjuyong.github.io\u002Findex.html)\u003C\u002Fli> \u003Cli>[Hujun Bao](https:\u002F\u002Fieeexplore.ieee.org\u002Fauthor\u002F37271755400)\u003C\u002Fli> \u003Cli>[Yong-Jin Liu](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=GNDtwWQAAAAJ)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_1c16c70541ee.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FTMM.2022.3207606) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fyiranran\u002FAudio-driven-TalkingFace-HeadPose?style=social)](https:\u002F\u002Fgithub.com\u002Fyiranran\u002FAudio-driven-TalkingFace-HeadPose) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2002.10137)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FJuyong\u002F3DFace), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FDeep3DFaceReconstruction), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FTEShHzqoTwk)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1gqcqTSAGAyj48n0fmApvSPG_43BzKP37) | 29.04.2020 |\n| Motion Supervised co-part Segmentation | A self-supervised deep learning method for co-part segmentation | \u003Cul>\u003Cli>[Aliaksandr Siarohin](https:\u002F\u002Faliaksandrsiarohin.github.io\u002Faliaksandr-siarohin-website\u002F)\u003C\u002Fli> \u003Cli>[Subhankar Roy](https:\u002F\u002Fgithub.com\u002Froysubhankar)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_4664349ec5ee.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FICPR48806.2021.9412520) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FAliaksandrSiarohin\u002Fmotion-cosegmentation?style=social)](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Fmotion-cosegmentation) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](http:\u002F\u002Farxiv.org\u002Fabs\u002F2004.03234)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAliaksandrSiarohin\u002Fvideo-preprocessing)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=RJ4Nj1wV5iA)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FAliaksandrSiarohin\u002Fmotion-cosegmentation\u002Fblob\u002Fmaster\u002Fpart_swap.ipynb) | 07.04.2020 |\n| Onsets and Frames | Onsets and Frames is an automatic music transcription framework with piano and drums models | \u003Cul>\u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli> \u003Cli>[Erich Elsen](https:\u002F\u002Fgithub.com\u002Fekelsen)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagenta\u002Fmagenta?style=social)](https:\u002F\u002Fgithub.com\u002Fmagenta\u002Fmagenta\u002Ftree\u002Fmain\u002Fmagenta\u002Fmodels\u002Fonsets_frames_transcription) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1710.11153), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.12247), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.00188)\u003C\u002Fli>\u003Cli>[blog post](http:\u002F\u002Fg.co\u002Fmagenta\u002Fonsets-frames)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fg.co\u002Fmagenta\u002Fmaestro-wave2midi2wave), [data](https:\u002F\u002Fmagenta.tensorflow.org\u002Fdatasets\u002Fe-gmd)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Fmagenta\u002Fonsets_frames_transcription\u002Fonsets_frames_transcription.ipynb) | 02.04.2020 |\n| FBA Matting | Low-cost modification to alpha matting networks to also predict the foreground and background colours | \u003Cul>\u003Cli>[Marco Forte](https:\u002F\u002Fgithub.com\u002FMarcoForte)\u003C\u002Fli> \u003Cli>[François Pitié](https:\u002F\u002Ffrancois.pitie.net\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FMarcoForte\u002FFBA_Matting?style=social)](https:\u002F\u002Fgithub.com\u002FMarcoForte\u002FFBA_Matting) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.07711)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMarcoForte\u002Fclosed-form-matting)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fspaces\u002Fleonelhs\u002FFBA-Matting)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fimage-matting-on-composition-1k?p=f-b-alpha-matting)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fdrive\u002F1Ut2szLBTxPejGHt_GYUkua21yUVWseOE) | 19.03.2020 |\n| BERT score | An automatic evaluation metric for text generation | [Tianyi Zhang](https:\u002F\u002Ftiiiger.github.io\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FTiiiger\u002Fbert_score?style=social)](https:\u002F\u002Fgithub.com\u002FTiiiger\u002Fbert_score) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.09675)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fbert-score\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FTiiiger\u002Fbert_score\u002Fblob\u002Fmaster\u002Fexample\u002FDemo.ipynb) | 05.03.2020 |\n| Deep Image Prior | Structure of a generator network is sufficient to capture a great deal of low-level image statistics prior to any learning | \u003Cul>\u003Cli>[Dmitry Ulyanov](https:\u002F\u002Fdmitryulyanov.github.io\u002Fabout)\u003C\u002Fli> \u003Cli>[Andrea Vedaldi](https:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vedaldi\u002F)\u003C\u002Fli> \u003Cli>[Victor Lempitsky](http:\u002F\u002Fsites.skoltech.ru\u002Fcompvision\u002Fmembers\u002Fvilem\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_443e7051c277.png)](https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs11263-020-01303-4) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FDmitryUlyanov\u002Fdeep-image-prior?style=social)](https:\u002F\u002Fgithub.com\u002FDmitryUlyanov\u002Fdeep-image-prior) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.10925)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fdmitryulyanov.github.io\u002Fdeep_image_prior)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002F7gls3j\u002Fr_deep_image_prior_deep_superresolution\u002F)\u003C\u002Fli>\u003Cli>[supmat](https:\u002F\u002Fbox.skoltech.ru\u002Findex.php\u002Fs\u002Fib52BOoV58ztuPM)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FDeep_image_prior)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-g1NsTuP1_I), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_BPJFFkxSbw), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F_WcGdXPdfjo), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FfCneY-7zFXE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FUYfPet5w_34), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIxIMvwkUsiQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3X5zfV5eQlY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FLzrUQtH43fY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FDmitryUlyanov\u002Fdeep-image-prior\u002Fblob\u002Fmaster\u002Factivation_maximization.ipynb) | 30.10.2019 |\n| ProxylessNAS | Directly learn the architectures for large-scale target tasks and target hardware platforms | \u003Cul>\u003Cli>[Han Cai](https:\u002F\u002Fhan-cai.github.io\u002F)\u003C\u002Fli> \u003Cli>[Ligeng Zhu](https:\u002F\u002Flzhu.me\u002F)\u003C\u002Fli> \u003Cli>[Song Han](https:\u002F\u002Fhanlab.mit.edu\u002Fsonghan)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmit-han-lab\u002Fproxylessnas?style=social)](https:\u002F\u002Fgithub.com\u002Fmit-han-lab\u002Fproxylessnas) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.00332), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.03494), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.08886)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Fpaper-proxylessnas-direct-neural-architecture-search-on-target-task-image-classification-73c35ebd8aed)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpt.svg\" alt=\"pt\" height=20\u002F>](https:\u002F\u002Fpytorch.org\u002Fhub\u002Fpytorch_vision_proxylessnas\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002FMachineLearning\u002Fcomments\u002Fa3a1xy\u002Fr_proxylessnas_direct_neural_architecture_search\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FX0aZmppnO1s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F6AeCIJH9eGI)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fpytorch\u002Fpytorch.github.io\u002Fblob\u002Fmaster\u002Fassets\u002Fhub\u002Fpytorch_vision_proxylessnas.ipynb) | 29.10.2019 |\n| Generating Piano Music with Transformer | This Colab notebook lets you play with pretrained Transformer models for piano music generation, based on the Music Transformer | \u003Cul>\u003Cli>[Ian Simon](https:\u002F\u002Fgithub.com\u002Fiansimon)\u003C\u002Fli> \u003Cli>[Anna Huang](https:\u002F\u002Fgithub.com\u002Fczhuang)\u003C\u002Fli> \u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli> \u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.03762), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.04281)\u003C\u002Fli>\u003Cli>[blog post](http:\u002F\u002Fg.co\u002Fmagenta\u002Fmusic-transformer)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Fmagenta\u002Fpiano_transformer\u002Fpiano_transformer.ipynb) | 16.09.2019 |\n| SSGAN | Self-Supervised GANs via Auxiliary Rotation Loss | \u003Cul>\u003Cli>[Ting Chen](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=KoXUMbsAAAAJ)\u003C\u002Fli> \u003Cli>[Xiaohua Zhai](https:\u002F\u002Fsites.google.com\u002Fview\u002Fxzhai)\u003C\u002Fli> \u003Cli>[Marvin Ritter](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=arcf5FgAAAAJ)\u003C\u002Fli> \u003Cli>[Mario Lučić](https:\u002F\u002Flucic.ai\u002F)\u003C\u002Fli> \u003Cli>[Neil Houlsby](https:\u002F\u002Fneilhoulsby.github.io\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_55478530ab65.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR.2019.01243) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fcompare_gan?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fcompare_gan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.11212), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.10337), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1807.04720), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1806.00035), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.04874), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.10340), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.01365), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.11212), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.02271)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fmodels\u002Fgoogle\u002Fcompare-gan)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science\u002Fself-supervised-gans-using-auxiliary-rotation-loss-60d8a929b556)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-oJWFcexolY)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fcompare_gan\u002Fblob\u002Fmaster\u002Fcolabs\u002Fssgan_demo.ipynb) | 20.06.2019 |\n| S3GAN | High-Fidelity Image Generation With Fewer Labels | \u003Cul>\u003Cli>[Mario Lucic](https:\u002F\u002Fresearch.google\u002Fpeople\u002Fmariolucic\u002F)\u003C\u002Fli> \u003Cli>[Michael Tschannen](https:\u002F\u002Fmitscha.github.io)\u003C\u002Fli> \u003Cli>[Marvin Ritter](https:\u002F\u002Fgithub.com\u002Fmritter0)\u003C\u002Fli> \u003Cli>[Xiaohua Zhai](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=sEkgK04AAAAJ)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Olivier Bachem](https:\u002F\u002Fresearch.google\u002Fpeople\u002Folivierbachem\u002F)\u003C\u002Fli> \u003Cli>[Sylvain Gelly](https:\u002F\u002Fwww.chessprogramming.org\u002FSylvain_Gelly)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fcompare_gan?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fcompare_gan) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.02271)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fcode\u002Fkerneler\u002Fgenerating-images-with-little-data-using-s3gan), [\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fmodels\u002Fgoogle\u002Fcompare-gan\u002FTensorFlow1\u002Fs3gan-20-128x128\u002F1)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Fhub\u002Ftutorials\u002Fs3gan_generation_with_tf_hub)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fcompare_gan\u002Fblob\u002Fmaster\u002Fcolabs\u002Fs3gan_demo.ipynb) | 10.06.2019 |\n| HMR | End-to-end framework for reconstructing a full 3D mesh of a human body from a single RGB image | \u003Cul>\u003Cli>[Angjoo Kanazawa](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~kanazawa\u002F)\u003C\u002Fli> \u003Cli>[Michael Black](https:\u002F\u002Fps.is.mpg.de\u002Fperson\u002Fblack)\u003C\u002Fli> \u003Cli>[David Jacobs](https:\u002F\u002Fwww.cs.umd.edu\u002F~djacobs\u002F)\u003C\u002Fli> \u003Cli>[Jitendra Malik](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~malik\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_abd81e31ab98.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FCVPR.2018.00744) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fakanazawa\u002Fhmr?style=social)](https:\u002F\u002Fgithub.com\u002Fakanazawa\u002Fhmr) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1712.06584)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocker.svg\" alt=\"docker\" height=20\u002F>](https:\u002F\u002Fhub.docker.com\u002Fr\u002Fdawars\u002Fhmr\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmattloper\u002Fchumpy), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FCMU-Perceptual-Computing-Lab\u002Fopenpose), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FMandyMo\u002Fpytorch_HMR), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Flayumi\u002Fhmr), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Frussoale\u002Fhmr2.0)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fakanazawa.github.io\u002Fhmr\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbmMV9aJKa-c)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FDene33\u002Fvideo_to_bvh\u002Fblob\u002Fmaster\u002Fvideo_to_bvh.ipynb) | 15.03.2019 |\n| GANSynth | This notebook is a demo GANSynth, which generates audio with Generative Adversarial Networks | [Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagenta\u002Fmagenta?style=social)](https:\u002F\u002Fgithub.com\u002Fmagenta\u002Fmagenta\u002Ftree\u002Fmain\u002Fmagenta\u002Fmodels\u002Fgansynth) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.08710), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.11096)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fstorage.googleapis.com\u002Fmagentadata\u002Fpapers\u002Fgansynth\u002Findex.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Fmagenta\u002Fgansynth\u002Fgansynth_demo.ipynb) | 25.02.2019 |\n| AmoebaNet | Regularized Evolution for Image Classifier Architecture Search | \u003Cul>\u003Cli>[Esteban Real](https:\u002F\u002Fwww.estebanreal.com\u002F)\u003C\u002Fli> \u003Cli>[Alok Aggarwal](https:\u002F\u002Fscryai.com\u002Fabout-us\u002Falok-aggarwal\u002F)\u003C\u002Fli> \u003Cli>[Yanping Huang](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=uEtBQScAAAAJ)\u003C\u002Fli> \u003Cli>[Quoc Le](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FQuoc_V._Le)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_2b2991845958.png)](https:\u002F\u002Fdoi.org\u002F10.1609\u002Faaai.v33i01.33014780) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle-research\u002Fgoogle-research?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle-research\u002Fgoogle-research\u002Ftree\u002Fmaster\u002Fevolution\u002Fregularized_evolution_algorithm) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.01548)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fsh-tsang.medium.com\u002Freading-amoebanet-regularized-evolution-for-image-classifier-architecture-search-image-278f5c077a4a)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fg557Nhg-f-k), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FOROmOKlcRxs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAYYhrfRPyww)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-research\u002Fgoogle-research\u002Fblob\u002Fmaster\u002Fevolution\u002Fregularized_evolution_algorithm\u002Fregularized_evolution.ipynb) | 25.10.2018 |\n| Latent Constraints | Conditional Generation from Unconditional Generative Models | \u003Cul>\u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli> \u003Cli>[Matthew Hoffman](http:\u002F\u002Fmatthewdhoffman.com\u002F)\u003C\u002Fli> \u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli>\u003C\u002Ful> | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1711.05772)\u003C\u002Fli>\u003Cli>[data](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FCelebA.html)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Flatent_constraints\u002Flatentconstraints.ipynb) | 27.11.2017 |\n| Performance RNN | This notebook shows you how to generate new performed compositions from a trained model | \u003Cul>\u003Cli>[Ian Simon](https:\u002F\u002Fgithub.com\u002Fiansimon)\u003C\u002Fli> \u003Cli>[Sageev Oore](https:\u002F\u002Fgithub.com\u002Fosageev)\u003C\u002Fli> \u003Cli>[Curtis Hawthorne](https:\u002F\u002Fgithub.com\u002Fcghawthorne)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagenta\u002Fmagenta?style=social)](https:\u002F\u002Fgithub.com\u002Fmagenta\u002Fmagenta\u002Ftree\u002Fmaster\u002Fmagenta\u002Fmodels\u002Fperformance_rnn) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fmagenta.tensorflow.org\u002Fperformance-rnn)\u003C\u002Fli>\u003Cli>[data](http:\u002F\u002Fwww.piano-e-competition.com\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Fmagenta\u002Fperformance_rnn\u002Fperformance_rnn.ipynb) | 11.07.2017 |\n| NSynth | This colab notebook has everything you need to upload your own sounds and use NSynth models to reconstruct and interpolate between them | \u003Cul>\u003Cli>[Jesse Engel](https:\u002F\u002Fgithub.com\u002Fjesseengel)\u003C\u002Fli> \u003Cli>[Cinjon Resnick](https:\u002F\u002Fgithub.com\u002Fcinjon)\u003C\u002Fli> \u003Cli>[Adam Roberts](https:\u002F\u002Fgithub.com\u002Fadarob)\u003C\u002Fli> \u003Cli>[Sander Dieleman](https:\u002F\u002Fbenanne.github.io\u002F)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Karen Simonyan](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=L7lMQkQAAAAJ)\u003C\u002Fli> \u003Cli>[Mohammad Norouzi](https:\u002F\u002Fnorouzi.github.io\u002F)\u003C\u002Fli> \u003Cli>[Douglas Eck](https:\u002F\u002Fgithub.com\u002Fdouglaseck)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Ftensorflow\u002Fmagenta?style=social)](https:\u002F\u002Fgithub.com\u002Ftensorflow\u002Fmagenta\u002Ftree\u002Fmaster\u002Fmagenta\u002Fmodels\u002Fnsynth) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1704.01279)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fmagenta.tensorflow.org\u002Fnsynth)\u003C\u002Fli>\u003Cli>[data](https:\u002F\u002Fmagenta.tensorflow.org\u002Fdatasets\u002Fnsynth)\u003C\u002Fli>\u003Cli>[tutorial](https:\u002F\u002Fmagenta.tensorflow.org\u002Fnsynth-fastgen)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=AaALLWQmCdI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=BOoSy-Pg8is)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fnotebooks\u002Fmagenta\u002Fnsynth\u002Fnsynth.ipynb) | 06.04.2017 |\n\n\u003C\u002Fdetails>\n\n\n\n## 教程\n\u003Cdetails>\n\u003Csummary>教程\u003C\u002Fsummary>\n\n| name | description | authors | links | colaboratory | update |\n|------|-------------|:--------|:------|:------------:|:------:|\n| Kornia | Library is composed by a subset of packages containing operators that can be inserted within neural networks to train models to perform image transformations, epipolar geometry, depth estimation, and low-level image processing such as filtering and edge detection that operate directly on tensors | \u003Cul>\u003Cli>[Edgar Riba](https:\u002F\u002Fgithub.com\u002Fedgarriba)\u003C\u002Fli> \u003Cli>[Dmytro Mishkin](https:\u002F\u002Fdmytro.ai\u002F)\u003C\u002Fli> \u003Cli>[Daniel Ponsa](https:\u002F\u002Fgithub.com\u002FDanielPonsa)\u003C\u002Fli> \u003Cli>[Ethan Rublee](https:\u002F\u002Fgithub.com\u002Fethanrublee)\u003C\u002Fli> \u003Cli>[Gary Bradski](https:\u002F\u002Fgithub.com\u002Fgarybradski)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_7f982a5488a1.png)](https:\u002F\u002Fdoi.org\u002F10.1109\u002FWACV45572.2020.9093363) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fkornia\u002Fkornia?style=social)](https:\u002F\u002Fgithub.com\u002Fkornia\u002Fkornia) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1910.02190)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fopencv.org\u002Fkornia-an-open-source-differentiable-computer-vision-library-for-pytorch\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002FHfnywwpBnD)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fkornia.readthedocs.io\u002Fen\u002Flatest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fkornia\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fslack.svg\" alt=\"slack\" height=20\u002F>](https:\u002F\u002Fjoin.slack.com\u002Ft\u002Fkornia\u002Fshared_invite\u002Fzt-csobk21g-2AQRi~X9Uu6PLMuUZdvfjA)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftwitter.svg\" alt=\"twitter\" height=20\u002F>](https:\u002F\u002Ftwitter.com\u002Fkornia_foss)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fkornia.github.io\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fchannel\u002FUCI1SE1Ij2Fast5BSKxoa7Ag), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3RmCYFhwclE), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAAZa-mXjYF0)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fkornia\u002Fkornia\u002Fblob\u002Fmaster\u002Fexamples\u002Faugmentation\u002Fkornia_augmentation.ipynb) | 28.11.2025 |\n| LM Evaluation Harness | Framework for few-shot evaluation of language models. | [Lintang Sutawika](https:\u002F\u002Flintang.sutawika.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FEleutherAI\u002Flm-evaluation-harness?style=social)](https:\u002F\u002Fgithub.com\u002FEleutherAI\u002Flm-evaluation-harness) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.14165)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdiscord.svg\" alt=\"discord\" height=20\u002F>](https:\u002F\u002Fdiscord.gg\u002Feleutherai)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FAutoGPTQ\u002FAutoGPTQ), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002FEleutherAI\u002Fgpt-neox), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fmicrosoft\u002FMegatron-DeepSpeed), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fvllm-project\u002Fvllm)\u003C\u002Fli>\u003Cli>[project](https:\u002F\u002Fwww.eleuther.ai\u002Fprojects\u002Flarge-language-model-evaluation)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FEleutherAI\u002Flm-evaluation-harness\u002Fblob\u002Fmain\u002Fexamples\u002Flm-eval-overview.ipynb) | 26.11.2025 |\n| Magenta RT | An open-weights live music model that allows you to interactively create, control and perform music in the moment | [Chris Donahue](https:\u002F\u002Fchrisdonahue.com\u002F) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fmagenta\u002Fmagenta-realtime?style=social)](https:\u002F\u002Fgithub.com\u002Fmagenta\u002Fmagenta-realtime) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2107.03312), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2208.12415), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2205.01917)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fmagenta.withgoogle.com\u002Fmagenta-realtime)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fai.google.dev\u002Fgemini-api\u002Fdocs\u002Fmusic-generation)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fgoogle\u002Fmagenta-realtime)\u003C\u002Fli>\u003Cli>[labs](https:\u002F\u002Flabs.google\u002Ffx\u002Ftools\u002Fmusic-fx-dj\u002Funsupported-country)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fdata-science-in-your-pocket\u002Fgoogle-magenta-realtime-ai-can-now-generate-songs-0cb3dbe01a00)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fmagenta\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FSVTuEdeepVs), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAe1Kz2zmh9M), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F6gVvsv3Va3s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FEg91A8sSWUM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3zkqFCaY3IQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fmagenta\u002Fmagenta-realtime\u002Fblob\u002Fmain\u002Fnotebooks\u002FMagenta_RT_Demo.ipynb) | 26.11.2025 |\n| SHAP | SHapley Additive exPlanations is a game theoretic approach to explain the output of any machine learning model | \u003Cul>\u003Cli>[Scott Lundberg](https:\u002F\u002Fscottlundberg.com\u002F)\u003C\u002Fli> \u003Cli>[Su-In Lee](https:\u002F\u002Fwww.cs.washington.edu\u002Fpeople\u002Ffaculty\u002Fsu-in-lee\u002F)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_a49ddcdcc789.png)](https:\u002F\u002Fdoi.org\u002F10.1038\u002Fs42256-019-0138-9) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fshap\u002Fshap?style=social)](https:\u002F\u002Fgithub.com\u002Fshap\u002Fshap) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.13972), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1704.02685), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1703.01365), [\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.03825)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fshap.readthedocs.io\u002Fen\u002Flatest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fneurips.svg\" alt=\"neurips\" height=20\u002F>](https:\u002F\u002Fproceedings.neurips.cc\u002Fpaper_files\u002Fpaper\u002F2017\u002Fhash\u002F8a20a8621978632d76c43dfd28b67767-Abstract.html)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fshap.readthedocs.io\u002Fen\u002Flatest\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FVB9uV-x0gtg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fd4PPMpdUCz8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fwjd1G5bu_TY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fd6PsRiEKTb8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F3BVU6nrfk4o), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FLsV-00K9W5I)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fshap\u002Fshap\u002Fblob\u002Fmaster\u002Fnotebooks\u002Foverviews\u002FAn%20introduction%20to%20explainable%20AI%20with%20Shapley%20values.ipynb#scrollTo=EAn8TeTgyA0Nhttps:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fshap\u002Fshap\u002Fblob\u002Fmaster\u002Fnotebooks\u002Foverviews\u002FAn%20introduction%20to%20explainable%20AI%20with%20Shapley%20values.ipynb) | 20.11.2025 |\n| Nano Banana | An image generation and editing model powered by generative artificial intelligence and developed by Google DeepMind | [Guillaume Vernade](https:\u002F\u002Fgithub.com\u002FGiom-V) | \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fblog.google\u002Ftechnology\u002Fai\u002Fnano-banana-pro\u002Fv)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fmodels\u002Fgemini-image\u002Fpro\u002F), [\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fmodels\u002Fgemini-image\u002Fflash\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fai.google.dev\u002Fgemini-api\u002Fdocs\u002Fimage-generation)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fgoogle-cloud\u002Fmy-experience-using-the-new-gemini-2-5-flash-image-8fbf79f00d76)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Faistudio.google.com\u002Fmodels\u002Fgemini-2-5-flash-image)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FNano_Banana)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8_GgeASwHwQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjtQiCJXOvdg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FRJ2eqkk_JxI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FIjWWnDZlezI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fm8Ve7hNl9P0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FexWEkRHmhKU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F5PiOEbnBDBs)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle-gemini\u002Fcookbook\u002Fblob\u002Fmain\u002Fquickstarts\u002FGet_Started_Nano_Banana.ipynb) | 20.11.2025 |\n| NeMo | A conversational AI toolkit built for researchers working on automatic speech recognition, natural language processing, and text-to-speech synthesis | \u003Cul>\u003Cli>[Oleksii Kuchaiev](http:\u002F\u002Fkuchaev.com\u002F)\u003C\u002Fli> \u003Cli>[Jason Li](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=V28bxDwAAAAJ)\u003C\u002Fli> \u003Cli>[Chip Huyen](https:\u002F\u002Fhuyenchip.com\u002F)\u003C\u002Fli> \u003Cli>[Oleksii Hrinchuk](https:\u002F\u002Fgithub.com\u002FAlexGrinch)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Ryan Leary](https:\u002F\u002Fgithub.com\u002Fryanleary)\u003C\u002Fli> \u003Cli>[Boris Ginsburg](https:\u002F\u002Fgithub.com\u002Fborisgin)\u003C\u002Fli> \u003Cli>[Samuel Kriman](https:\u002F\u002Fgithub.com\u002Fsam1373)\u003C\u002Fli> \u003Cli>[Stanislav Beliaev](https:\u002F\u002Fgithub.com\u002Fstasbel)\u003C\u002Fli> \u003Cli>[Vitaly Lavrukhin](https:\u002F\u002Fgithub.com\u002Fvsl9)\u003C\u002Fli> \u003Cli>[Jack Cook](https:\u002F\u002Fjackcook.com\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FNVIDIA\u002FNeMo?style=social)](https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNeMo) \u003Cul>\u003Cli>[project](https:\u002F\u002Fdocs.nvidia.com\u002Fdeeplearning\u002Fnemo\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FwBgpMf_KQVw)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FNVIDIA\u002FNeMo\u002Fblob\u002Fmaster\u002Ftutorials\u002F00_NeMo_Primer.ipynb) | 17.11.2025 |\n| PyTerrier | A Python framework for performing information retrieval experiments | \u003Cul>\u003Cli>[Craig Macdonald](https:\u002F\u002Fwww.dcs.gla.ac.uk\u002F~craigm\u002F)\u003C\u002Fli> \u003Cli>[Nicola Tonellotto](https:\u002F\u002Fgithub.com\u002Ftonellotto)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_readme_346a1cde8fcc.png)](https:\u002F\u002Fdoi.org\u002F10.1145\u002F3459637.3482013) [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fterrier-org\u002Fpyterrier?style=social)](https:\u002F\u002Fgithub.com\u002Fterrier-org\u002Fpyterrier) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2007.14271)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fpyterrier.readthedocs.io)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrier-org\u002Fecir2021tutorial), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_ance), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_colbert), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_pisa), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_t5), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_doc2query), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fterrierteam\u002Fpyterrier_deepct)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fterrier-org\u002Fpyterrier\u002Fblob\u002Fmaster\u002Fexamples\u002Fnotebooks\u002Fnon_en_retrieval.ipynb) | 13.11.2025 |\n| Transfer learning and fine-tuning | You will learn how to classify images of cats and dogs by using transfer learning from a pre-trained network | [François Chollet](https:\u002F\u002Ffchollet.com\u002F) | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fpwc.svg\" alt=\"pwc\" height=20\u002F>](https:\u002F\u002Fpaperswithcode.com\u002Ftask\u002Ftransfer-learning)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Ftf.svg\" alt=\"tf\" height=20\u002F>](https:\u002F\u002Fwww.tensorflow.org\u002Ftutorials\u002Fimages\u002Ftransfer_learning)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fwiki.svg\" alt=\"wiki\" height=20\u002F>](https:\u002F\u002Fen.wikipedia.org\u002Fwiki\u002FTransfer_learning)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Ftensorflow\u002Fdocs\u002Fblob\u002Fmaster\u002Fsite\u002Fen\u002Ftutorials\u002Fimages\u002Ftransfer_learning.ipynb) | 11.11.2025 |\n| Datasets | A Community Library for Natural Language Processing | \u003Cul>\u003Cli>[Quentin Lhoest](https:\u002F\u002Fgithub.com\u002Flhoestq)\u003C\u002Fli> \u003Cli>[Albert Villanova](https:\u002F\u002Falbertvillanova.github.io\u002F)\u003C\u002Fli> \u003Cli>[Yacine Jernite](https:\u002F\u002Fyjernite.github.io\u002F)\u003C\u002Fli> \u003Cli>[Abhishek Thakur](https:\u002F\u002Fgithub.com\u002Fabhishekkrthakur)\u003C\u002Fli>\u003Cdetails>\u003Csummary>others\u003C\u002Fsummary>\u003Cli>[Patrick von Platen](https:\u002F\u002Fgithub.com\u002Fpatrickvonplaten)\u003C\u002Fli> \u003Cli>[Suraj Patil](https:\u002F\u002Fgithub.com\u002Fpatil-suraj)\u003C\u002Fli> \u003Cli>[Julien Chaumond](https:\u002F\u002Fgithub.com\u002Fjulien-c)\u003C\u002Fli> \u003Cli>[Mariama Dramé](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=0pwfXH0AAAAJ)\u003C\u002Fli> \u003Cli>[Julien Plu](https:\u002F\u002Fjplu.github.io\u002F)\u003C\u002Fli> \u003Cli>[Lewis Tunstall](https:\u002F\u002Flewtun.github.io\u002Fblog\u002F)\u003C\u002Fli> \u003Cli>[Joe Davison](https:\u002F\u002Fjoeddav.github.io\u002F)\u003C\u002Fli> \u003Cli>[Mario Šaško](https:\u002F\u002Fgithub.com\u002Fmariosasko)\u003C\u002Fli> \u003Cli>[Gunjan Chhablani](https:\u002F\u002Fgchhablani.github.io\u002F)\u003C\u002Fli> \u003Cli>[Bhavitvya Malik](https:\u002F\u002Fgithub.com\u002Fbhavitvyamalik)\u003C\u002Fli> \u003Cli>[Simon Brandeis](https:\u002F\u002Fgithub.com\u002FSBrandeis)\u003C\u002Fli> \u003Cli>[Teven Le Scao](https:\u002F\u002Fgithub.com\u002FTevenLeScao)\u003C\u002Fli> \u003Cli>[Victor Sanh](https:\u002F\u002Fgithub.com\u002FVictorSanh)\u003C\u002Fli> \u003Cli>[Canwen Xu](https:\u002F\u002Fwww.canwenxu.net\u002F)\u003C\u002Fli> \u003Cli>[Nicolas Patry](https:\u002F\u002Fgithub.com\u002FNarsil)\u003C\u002Fli> \u003Cli>[Angelina McMillan-Major](https:\u002F\u002Fgithub.com\u002Fmcmillanmajora)\u003C\u002Fli> \u003Cli>[Philipp Schmid](https:\u002F\u002Fwww.philschmid.de\u002F)\u003C\u002Fli> \u003Cli>[Sylvain Gugger](https:\u002F\u002Fgithub.com\u002Fsgugger)\u003C\u002Fli> \u003Cli>[Clément Delangue](https:\u002F\u002Fscholar.google.com\u002Fcitations?user=bRMboT8AAAAJ)\u003C\u002Fli> \u003Cli>[Théo Matussière](https:\u002F\u002Ftheo.matussie.re\u002F)\u003C\u002Fli> \u003Cli>[Lysandre Debut](http:\u002F\u002Flysand.re\u002F)\u003C\u002Fli> \u003Cli>[Stas Bekman](https:\u002F\u002Fstasosphere.com\u002Fmachine-learning\u002F)\u003C\u002Fli> \u003Cli>[Pierric Cistac](https:\u002F\u002Fgithub.com\u002FPierrci)\u003C\u002Fli> \u003Cli>[Thibault Goehringer](https:\u002F\u002Fgithub.com\u002Fbeurkinger)\u003C\u002Fli> \u003Cli>[Victor Mustar](https:\u002F\u002Fgithub.com\u002Fgary149)\u003C\u002Fli> \u003Cli>[François Lagunas](https:\u002F\u002Fgithub.com\u002Fmadlag)\u003C\u002Fli> \u003Cli>[Alexander Rush](https:\u002F\u002Frush-nlp.com\u002F)\u003C\u002Fli> \u003Cli>[Thomas Wolf](https:\u002F\u002Fthomwolf.io\u002F)\u003C\u002Fli>\u003C\u002Ful>\u003C\u002Fdetails> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fhuggingface\u002Fdatasets?style=social)](https:\u002F\u002Fgithub.com\u002Fhuggingface\u002Fdatasets) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2109.02846)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdocs\u002Fdatasets)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fhf.svg\" alt=\"hf\" height=20\u002F>](https:\u002F\u002Fhuggingface.co\u002Fdatasets)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fkaggle.svg\" alt=\"kaggle\" height=20\u002F>](https:\u002F\u002Fwww.kaggle.com\u002Fcode\u002Fnbroad\u002Fintro-to-hugging-face-datasets)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fdatasets\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FuaIJ96syPnM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fhuggingface\u002Fnotebooks\u002Fblob\u002Fmain\u002Fdatasets_doc\u002Fen\u002Fquickstart.ipynb) | 10.11.2025 |\n| Agent Starter Pack | Collection of production-ready Generative AI Agent templates built for Google Cloud | [Kristopher Overholt](https:\u002F\u002Fgithub.com\u002Fkoverholt) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FGoogleCloudPlatform\u002Fagent-starter-pack?style=social)](https:\u002F\u002Fgithub.com\u002FGoogleCloudPlatform\u002Fagent-starter-pack) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fgoogle-cloud\u002Fagentic-rag-with-bigquery-dataframes-and-agent-starter-pack-743553adf997), [\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002Fgoogle-cloud\u002Fgenai-app-starter-pack-now-with-rag-pattern-vertex-ai-search-fb81bf61bcd5)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fagent-starter-pack\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fgooglecloud\u002Fcomments\u002F1j49uz7\u002Fagent_starter_pack_build_deploy_genai_agents_on\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FjHt-ZVD660g), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FkwRG7cnqSu0), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=yIRIT_EtALs&t=235s), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLIivdWyY5sqLRCzKJyixrIDPQKwU6XHpn)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGoogleCloudPlatform\u002Fgenerative-ai\u002Fblob\u002Fmain\u002Fgemini\u002Fagent-engine\u002Fintro_agent_engine.ipynb) | 06.11.2025 |\n| Google Cloud Text-to-Speech | Enables easy integration of Google text recognition technologies into developer applications | \u003Cul>\u003Cli>[Holt Skinner](https:\u002F\u002Fgithub.com\u002Fholtskinner)\u003C\u002Fli> \u003Cli>[Ivan Nardini](https:\u002F\u002Fgithub.com\u002Finardini)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogleapis\u002Fgoogle-cloud-python?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogleapis\u002Fgoogle-cloud-python\u002Ftree\u002Fmain\u002Fpackages\u002Fgoogle-cloud-texttospeech) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fcloud.google.com\u002Fpython\u002Fdocs\u002Freference\u002Ftexttospeech\u002Flatest\u002Fsummary_overview)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgoogle-cloud-texttospeech\u002F)\u003C\u002Fli>\u003Cli>[website](https:\u002F\u002Fcloud.google.com\u002Ftext-to-speech)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGVPWz-nhJhg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fwzp9dfVpeeg), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FNaAB9ogyB-U), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fz8g3XM16eRM), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FDCsVRs1-Hn8), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FSPcFViKU_xU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FlKra6E_tp5U), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FdOlV_oD_dr8)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGoogleCloudPlatform\u002Fgenerative-ai\u002Fblob\u002Fmain\u002Faudio\u002Fspeech\u002Fgetting-started\u002Fget_started_with_chirp_3_hd_voices.ipynb) | 06.11.2025 |\n| Imagen 4 | Text-to-image model, with photorealistic images, near real-time speed, and sharper clarity | [Katie Nguyen](https:\u002F\u002Fgithub.com\u002Fkatiemn) | \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fdevelopers.googleblog.com\u002Fen\u002Fimagen-4-now-available-in-the-gemini-api-and-google-ai-studio\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fmodels\u002Fimagen\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgoogle-genai\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-f4I1_27SwI), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FAzwg982tpvM)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGoogleCloudPlatform\u002Fgenerative-ai\u002Fblob\u002Fmain\u002Fvision\u002Fgetting-started\u002Fimagen4_image_generation.ipynb) | 06.11.2025 |\n| Lyria 2 | Delivers high-fidelity music and professional-grade audio, capturing subtle nuances across a range of genres and intricate compositions | [Katie Nguyen](https:\u002F\u002Fgithub.com\u002Fkatiemn) | \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Fmodels\u002Flyria\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fcloud.google.com\u002Fvertex-ai\u002Fgenerative-ai\u002Fdocs\u002Fmusic\u002Fgenerate-music)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fsingularity\u002Fcomments\u002F1ku5tdw\u002Flyria_2_googles_new_ai_music_generator_sounds\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FbiJJEbVjvko)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGoogleCloudPlatform\u002Fgenerative-ai\u002Fblob\u002Fmain\u002Faudio\u002Fmusic\u002Fgetting-started\u002Flyria2_music_generation.ipynb) | 06.11.2025 |\n| Vertex AI | Search brings together the power of deep information retrieval, state-of-the-art natural language processing, and the latest in LLM processing to understand user intent and return the most relevant results for the user | [Megha Agarwal](https:\u002F\u002Fgithub.com\u002Fagarwal22megha) | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogleapis\u002Fpython-aiplatform?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogleapis\u002Fpython-aiplatform) \u003Cul>\u003Cli>[\u003Cimg src=\"images\u002Farxiv.svg\" alt=\"arxiv\" height=20\u002F>](https:\u002F\u002Farxiv.org\u002Fabs\u002F2407.16833)\u003C\u002Fli>\u003Cli>[blog post](https:\u002F\u002Fblog.google\u002Fproducts\u002Fsearch\u002Fimproving-search-next-20-years\u002Fv)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdeepmind.svg\" alt=\"deepmind\" height=20\u002F>](https:\u002F\u002Fdeepmind.google\u002Ftechnologies\u002Fgemini\u002Fpro\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fcloud.google.com\u002Fgenerative-ai-app-builder\u002Fdocs\u002Fenterprise-search-introduction)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgoogle-cloud-aiplatform\u002F), [\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgoogle-cloud-discoveryengine\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLIivdWyY5sqJ1YuMdGjRwJ3fFYZ_vWQ62), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLIivdWyY5sqJAyUJbbsc8ZyGLNT4isnuB), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FHD_xreaLKb4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F-3Olw-C4FN4), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FXVO3zsHdvio)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002FGoogleCloudPlatform\u002Fgenerative-ai\u002Fblob\u002Fmain\u002Fsearch\u002Fvertexai-search-options\u002Fvertexai_search_options.ipynb) | 06.11.2025 |\n| ADK | Collection provides ready-to-use agents built on top of the Agent Development Kit, designed to accelerate your development process | \u003Cul>\u003Cli>[Equious](https:\u002F\u002Fgithub.com\u002FEquious)\u003C\u002Fli> \u003Cli>[Ankur Sharma](https:\u002F\u002Fgithub.com\u002Fankursharmas)\u003C\u002Fli>\u003C\u002Ful> | [![](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fgoogle\u002Fadk-samples?style=social)](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fadk-samples) \u003Cul>\u003Cli>[blog post](https:\u002F\u002Fdevelopers.googleblog.com\u002Fen\u002Fagent-development-kit-easy-to-build-multi-agent-applications\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fgoogle.github.io\u002Fadk-docs\u002F), [\u003Cimg src=\"images\u002Fdocs.svg\" alt=\"docs\" height=20\u002F>](https:\u002F\u002Fdocs.cloud.google.com\u002Fagent-builder\u002Fagent-development-kit\u002Foverview)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fadk-go), [\u003Cimg src=\"images\u002Fgit.svg\" alt=\"git\" height=20\u002F>](https:\u002F\u002Fgithub.com\u002Fgoogle\u002Fadk-java)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fmedium.svg\" alt=\"medium\" height=20\u002F>](https:\u002F\u002Fmedium.com\u002F@d3xvn\u002Fexploring-googles-agent-development-kit-adk-71a27a609920)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fpypi.svg\" alt=\"pypi\" height=20\u002F>](https:\u002F\u002Fpypi.org\u002Fproject\u002Fgoogle-adk\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Freddit.svg\" alt=\"reddit\" height=20\u002F>](https:\u002F\u002Fwww.reddit.com\u002Fr\u002Fagentdevelopmentkit\u002F)\u003C\u002Fli>\u003Cli>[\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fwww.youtube.com\u002Fplaylist?list=PLOU2XLYxmsIIAPgM8FmtEcFTXLLzmh4DK), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FP4VFL9nIaIA), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F44C8u0CDtSo), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FG9wnpfW6lZY), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002F8rlNdKywldQ), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FBiP4tKZKTvU), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002Fcz2pKLPw994), [\u003Cimg src=\"images\u002Fyt.svg\" alt=\"yt\" height=20\u002F>](https:\u002F\u002Fyoutu.be\u002FGeo8LzCHoMQ)\u003C\u002Fli>\u003C\u002Ful> | [![Open In Colab](images\u002Fcolab.svg)](https:\u002F\u002Fcolab.research.google.com\u002Fgithub\u002Fgoogle\u002Fadk-samples\u002Fblob\u002Fmain\u002Fpython\u002Fnotebooks\u002Fevaluation\u002Fuser_simulation_in_adk_evals.ipynb) | 05.11.2025 |\n| TRL | Set of tools to train trans","# awesome-colab-notebooks 快速上手指南\n\n`awesome-colab-notebooks` 是一个精选的机器学习实验 Colab Notebook 集合，涵盖了从热门开源项目、学术论文复现到实用 Python 包的各种资源。本指南将帮助你快速利用这些资源在 Google Colab 环境中进行开发和实验。\n\n## 环境准备\n\n由于该工具本质上是 **Google Colab Notebook 的索引集合**，无需在本地安装复杂的环境。你只需要满足以下条件：\n\n1.  **Google 账号**：用于登录 Google Colab。\n2.  **浏览器**：推荐使用 Chrome 或 Edge 浏览器。\n3.  **网络环境**：\n    *   需要能够访问 `github.com` 和 `colab.research.google.com`。\n    *   **国内开发者提示**：由于 Google 服务在中国大陆无法直接访问，你需要配置科学上网环境（全局代理模式效果最佳），或者使用支持挂载 GitHub 的国内替代平台（如 AutoDL、Kaggle 等，但部分链接可能需要手动导入）。\n\n**前置依赖**：\n*   无本地依赖。所有依赖库（如 `torch`, `tensorflow`, `transformers` 等）将在每个具体的 Notebook 运行时自动安装。\n\n## 安装步骤\n\n本项目不需要传统的“安装”过程。使用流程如下：\n\n1.  **访问仓库主页**：\n    打开浏览器访问项目 GitHub 页面：\n    ```text\n    https:\u002F\u002Fgithub.com\u002Famrzv\u002Fawesome-colab-notebooks\n    ```\n\n2.  **选择目标资源**：\n    在 README 中浏览以下分类找到你感兴趣的内容：\n    *   **Trending (Repositories)**：热门开源项目（如 `agent-starter-pack`, `Qwen3-Omni`, `langgraph`）。\n    *   **Papers**：附带论文的实现代码（如 `LLaMA Factory`, `Gaussian Splatting`）。\n    *   **Packages**：常用 AI 包的演示（如 `sam3`, `sglang`）。\n    *   **Courses**：系统性课程（如 `LLM Engineering Essentials`）。\n\n3.  **启动 Colab 环境**：\n    点击对应项目链接进入其 GitHub 页面，寻找 **\"Open in Colab\"** 按钮，或者直接点击 README 表格中的 `colaboratory` 列链接（如果有）。\n\n    *如果只有 GitHub 链接，手动在 Colab 中打开的方法：*\n    复制目标 `.ipynb` 文件的 GitHub URL，然后在 Google Colab 首页选择 `File` -> `Open notebook` -> `GitHub` 标签页，粘贴链接并打开。\n\n## 基本使用\n\n以下以运行一个典型的深度学习模型（例如从列表中选择一个 BERT 微调或图像分割项目）为例：\n\n### 1. 打开 Notebook\n点击选定的 Colab 链接，页面加载后你会看到代码单元格。\n\n### 2. 设置运行时类型（重要）\n为了获得 GPU 加速，请检查运行时设置：\n*   点击菜单栏的 **Runtime (运行时)** > **Change runtime type (更改运行时类型)**。\n*   在 **Hardware accelerator (硬件加速器)** 下拉菜单中选择 **GPU** (推荐 T4) 或 **TPU**。\n*   点击 **Save**。\n\n### 3. 执行代码\n按顺序执行单元格。通常第一个单元格是安装依赖，后续是加载数据和训练\u002F推理。\n\n**示例：安装依赖并运行（伪代码逻辑，具体视 Notebook 而定）**\n\n```python\n# 单元格 1: 安装必要的库 (通常在 Notebook 开头已提供)\n!pip install transformers datasets accelerate\n\n# 单元格 2: 导入库并加载模型\nfrom transformers import pipeline\n\nclassifier = pipeline(\"sentiment-analysis\")\nresult = classifier(\"I love using Colab for AI experiments!\")\nprint(result)\n```\n\n### 4. 挂载 Google Drive（可选，用于保存结果）\n如果需要保存训练好的模型或输出文件，建议挂载云盘：\n\n```python\nfrom google.colab import drive\ndrive.mount('\u002Fcontent\u002Fdrive')\n```\n\n### 5. 国内加速方案（针对依赖下载失败）\n如果在 Colab 中下载 Python 包速度过慢或被阻断，可以在安装命令中临时指定清华源或阿里源：\n\n```bash\n# 原命令示例\n!pip install torch torchvision\n\n# 修改为使用清华源加速\n!pip install torch torchvision -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple\n```\n\n*注意：部分 Colab 环境可能对国内镜像源的网络策略有限制，如遇连接超时，请尝试切换回默认源或使用代理设置。*","一名计算机视觉研究生正急于复现最新的交互式图像分割论文（如 ritm-interactive-segmentation），以验证其在自定义数据集上的效果，但受限于本地显卡显存不足且环境配置复杂。\n\n### 没有 awesome-colab-notebooks 时\n- **环境配置噩梦**：需手动在本地安装 CUDA、cuDNN 及各类依赖库，常因版本冲突导致“依赖地狱”，耗费数天仍无法跑通代码。\n- **硬件门槛高企**：本地消费级显卡显存不足以加载大模型，强行运行会导致内存溢出（OOM），迫使研究者不得不申请昂贵的云服务器或排队使用实验室集群。\n- **复现路径迷茫**：面对 GitHub 上零散的官方仓库，缺乏经过调试的标准化入口，难以快速定位可执行的演示代码，大量时间浪费在排查基础报错上。\n- **实验迭代缓慢**：每次调整超参数或更换数据集都需要重新配置环境，无法实现“开箱即用”的快速验证，严重拖慢科研进度。\n\n### 使用 awesome-colab-notebooks 后\n- **一键启动实验**：直接点击集合中对应的 Colab 链接（如 ritm-interactive-segmentation 笔记本），无需本地安装任何环境，浏览器即可调用云端 GPU 立即运行。\n- **免费算力支持**：利用 Google Colab 提供的免费 T4\u002FP100 GPU 资源，轻松突破本地硬件限制，流畅运行原本需要高端显卡才能负载的深度学习模型。\n- **标准化复现流程**：借助社区维护的成熟笔记，直接获取已预装好依赖、打通数据加载与推理流程的代码，将复现时间从数天缩短至几分钟。\n- **敏捷迭代验证**：可在云端直接修改代码单元格测试不同参数或上传私有数据，实时查看分割结果，极大提升了算法验证与调优的效率。\n\nawesome-colab-notebooks 通过将复杂的工程配置封装为云端即点即用的实验模板，让研究者能从繁琐的环境搭建中解脱，专注于核心算法的创新与验证。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Famrzv_awesome-colab-notebooks_9ea152c3.png","amrzv","Alexey Morozov","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Famrzv_99eb00c7.jpg",null,"@SoundHound","EU","amrzv.github.io","https:\u002F\u002Fgithub.com\u002Famrzv",[81],{"name":82,"color":83,"percentage":84},"Python","#3572A5",100,1625,277,"2026-04-06T06:42:39","MIT",1,"未说明","未说明 (该项目为 Colab 笔记本集合，具体 GPU 需求取决于所选的单个笔记本，通常由 Google Colab 免费提供 T4\u002FP100\u002FA100 等)","未说明 (取决于具体实验，Colab 免费版通常提供约 12GB RAM)",{"notes":94,"python":95,"dependencies":96},"该项目本身不是一个单一的可安装工具，而是一个精选的机器学习 Colab 笔记本列表。运行环境完全依赖于 Google Colab 云端平台，无需本地配置操作系统或驱动。具体的库依赖和硬件需求（如显存大小）因用户选择运行的特定笔记本（如 LLaMA Factory, Gaussian Splatting 等）而异。建议直接在浏览器中打开链接使用，或根据特定笔记本的说明在本地复现环境。","未说明 (由 Google Colab 环境决定，通常为 Python 3.10+)",[97,98,99,100,101,102,103,104,105,106],"torch","transformers","tensorflow","jax","langgraph","sglang","swanlab","opik","accelerate","diffusers",[14],[109,110,111,112,99,113,114,115,116,117,118,119,120,121,122,123],"colab-notebooks","machine-learning","deep-learning","generative-adversarial-network","tensorflow-tutorials","jupyter-notebooks","google-colaboratory","google-colab","google-colab-tutorial","google-colabs","google-colab-notebooks","google-colab-notebook","deep-neural-networks","pytorch","cnn","2026-03-27T02:49:30.150509","2026-04-07T22:59:44.471487",[127,132],{"id":128,"question_zh":129,"answer_zh":130,"source_url":131},23140,"RuGPT3XL Colab 笔记本在运行时报错（如 deep speed、sparse attention ops 或 logits 相关错误）怎么办？","尝试重置运行时环境。具体步骤为：点击菜单栏的 \"Runtime\"（运行时） -> 选择 \"Reset all runtimes\"（重置所有运行时），然后重新运行笔记本。维护者测试表明这通常能解决此类 CUDA 或环境缓存导致的错误。","https:\u002F\u002Fgithub.com\u002Famrzv\u002Fawesome-colab-notebooks\u002Fissues\u002F3",{"id":133,"question_zh":134,"answer_zh":135,"source_url":136},23141,"如何向该仓库提交新的 Colab 笔记本示例？需要遵循什么格式？","提交时请提供包含名称、描述和链接的 Markdown 表格行，以便维护者将其添加到列表中。格式如下：\n| name | description | link |\n|------|-------------|:------------:|\n例如：\n| Playing CartPole with the Actor-Critic Method | This tutorial demonstrates how to implement the Actor-Critic method using TensorFlow... | https:\u002F\u002Fwww.tensorflow.org\u002Ftutorials\u002Freinforcement_learning\u002Factor_critic |","https:\u002F\u002Fgithub.com\u002Famrzv\u002Fawesome-colab-notebooks\u002Fissues\u002F1",[]]