[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-timzhang642--3D-Machine-Learning":3,"tool-timzhang642--3D-Machine-Learning":64},[4,17,27,35,43,56],{"id":5,"name":6,"github_repo":7,"description_zh":8,"stars":9,"difficulty_score":10,"last_commit_at":11,"category_tags":12,"status":16},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。",162132,3,"2026-04-05T11:01:52",[13,14,15],"开发框架","图像","Agent","ready",{"id":18,"name":19,"github_repo":20,"description_zh":21,"stars":22,"difficulty_score":23,"last_commit_at":24,"category_tags":25,"status":16},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",138956,2,"2026-04-05T11:33:21",[13,15,26],"语言模型",{"id":28,"name":29,"github_repo":30,"description_zh":31,"stars":32,"difficulty_score":23,"last_commit_at":33,"category_tags":34,"status":16},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。",107662,"2026-04-03T11:11:01",[13,14,15],{"id":36,"name":37,"github_repo":38,"description_zh":39,"stars":40,"difficulty_score":23,"last_commit_at":41,"category_tags":42,"status":16},3704,"NextChat","ChatGPTNextWeb\u002FNextChat","NextChat 是一款轻量且极速的 AI 助手，旨在为用户提供流畅、跨平台的大模型交互体验。它完美解决了用户在多设备间切换时难以保持对话连续性，以及面对众多 AI 模型不知如何统一管理的痛点。无论是日常办公、学习辅助还是创意激发，NextChat 都能让用户随时随地通过网页、iOS、Android、Windows、MacOS 或 Linux 端无缝接入智能服务。\n\n这款工具非常适合普通用户、学生、职场人士以及需要私有化部署的企业团队使用。对于开发者而言，它也提供了便捷的自托管方案，支持一键部署到 Vercel 或 Zeabur 等平台。\n\nNextChat 的核心亮点在于其广泛的模型兼容性，原生支持 Claude、DeepSeek、GPT-4 及 Gemini Pro 等主流大模型，让用户在一个界面即可自由切换不同 AI 能力。此外，它还率先支持 MCP（Model Context Protocol）协议，增强了上下文处理能力。针对企业用户，NextChat 提供专业版解决方案，具备品牌定制、细粒度权限控制、内部知识库整合及安全审计等功能，满足公司对数据隐私和个性化管理的高标准要求。",87618,"2026-04-05T07:20:52",[13,26],{"id":44,"name":45,"github_repo":46,"description_zh":47,"stars":48,"difficulty_score":23,"last_commit_at":49,"category_tags":50,"status":16},2268,"ML-For-Beginners","microsoft\u002FML-For-Beginners","ML-For-Beginners 是由微软推出的一套系统化机器学习入门课程，旨在帮助零基础用户轻松掌握经典机器学习知识。这套课程将学习路径规划为 12 周，包含 26 节精炼课程和 52 道配套测验，内容涵盖从基础概念到实际应用的完整流程，有效解决了初学者面对庞大知识体系时无从下手、缺乏结构化指导的痛点。\n\n无论是希望转型的开发者、需要补充算法背景的研究人员，还是对人工智能充满好奇的普通爱好者，都能从中受益。课程不仅提供了清晰的理论讲解，还强调动手实践，让用户在循序渐进中建立扎实的技能基础。其独特的亮点在于强大的多语言支持，通过自动化机制提供了包括简体中文在内的 50 多种语言版本，极大地降低了全球不同背景用户的学习门槛。此外，项目采用开源协作模式，社区活跃且内容持续更新，确保学习者能获取前沿且准确的技术资讯。如果你正寻找一条清晰、友好且专业的机器学习入门之路，ML-For-Beginners 将是理想的起点。",84991,"2026-04-05T10:45:23",[14,51,52,53,15,54,26,13,55],"数据工具","视频","插件","其他","音频",{"id":57,"name":58,"github_repo":59,"description_zh":60,"stars":61,"difficulty_score":10,"last_commit_at":62,"category_tags":63,"status":16},3128,"ragflow","infiniflow\u002Fragflow","RAGFlow 是一款领先的开源检索增强生成（RAG）引擎，旨在为大语言模型构建更精准、可靠的上下文层。它巧妙地将前沿的 RAG 技术与智能体（Agent）能力相结合，不仅支持从各类文档中高效提取知识，还能让模型基于这些知识进行逻辑推理和任务执行。\n\n在大模型应用中，幻觉问题和知识滞后是常见痛点。RAGFlow 通过深度解析复杂文档结构（如表格、图表及混合排版），显著提升了信息检索的准确度，从而有效减少模型“胡编乱造”的现象，确保回答既有据可依又具备时效性。其内置的智能体机制更进一步，使系统不仅能回答问题，还能自主规划步骤解决复杂问题。\n\n这款工具特别适合开发者、企业技术团队以及 AI 研究人员使用。无论是希望快速搭建私有知识库问答系统，还是致力于探索大模型在垂直领域落地的创新者，都能从中受益。RAGFlow 提供了可视化的工作流编排界面和灵活的 API 接口，既降低了非算法背景用户的上手门槛，也满足了专业开发者对系统深度定制的需求。作为基于 Apache 2.0 协议开源的项目，它正成为连接通用大模型与行业专有知识之间的重要桥梁。",77062,"2026-04-04T04:44:48",[15,14,13,26,54],{"id":65,"github_repo":66,"name":67,"description_en":68,"description_zh":69,"ai_summary_zh":69,"readme_en":70,"readme_zh":71,"quickstart_zh":72,"use_case_zh":73,"hero_image_url":74,"owner_login":75,"owner_name":76,"owner_avatar_url":77,"owner_bio":78,"owner_company":79,"owner_location":79,"owner_email":80,"owner_twitter":81,"owner_website":82,"owner_url":83,"languages":79,"stars":84,"forks":85,"last_commit_at":86,"license":79,"difficulty_score":87,"env_os":88,"env_gpu":89,"env_ram":89,"env_deps":90,"category_tags":93,"github_topics":94,"view_count":23,"oss_zip_url":79,"oss_zip_packed_at":79,"status":16,"created_at":105,"updated_at":106,"faqs":107,"releases":108},3532,"timzhang642\u002F3D-Machine-Learning","3D-Machine-Learning","A resource repository for 3D machine learning","3D-Machine-Learning 是一个专注于三维机器学习领域的开源资源库，旨在汇集计算机视觉、图形学与机器学习交叉学科的前沿成果。面对该领域技术迭代快、论文数量庞大且数据表示形式复杂（如多视图图像、体素、点云、网格及图元等）的挑战，它提供了一套系统化的知识整理方案。\n\n该项目不仅分类整理了从基础课程、权威数据集到最新研究论文的海量资料，还覆盖了三维姿态估计、物体检测、语义分割、几何重建及场景理解等核心任务。其独特亮点在于通过直观的图标体系区分不同的三维数据表示形式，帮助读者快速定位所需内容；同时推荐利用关联图谱工具可视化学术脉络，极大提升了文献调研效率。此外，项目还建立了全球协作的社区环境，促进知识共享与合作。\n\n无论是希望入门三维深度学习的学生、需要追踪最新算法的研究人员，还是寻求工程落地的开发者，都能从中获得宝贵的学习路径与参考资料。它将零散的学术资源转化为结构清晰的知识体系，是探索三维智能技术不可或缺的指南针。","3D Machine Learning\n\nIn recent years, tremendous amount of progress is being made in the field of 3D Machine Learning, which is an interdisciplinary field that fuses computer vision, computer graphics and machine learning. This repo is derived from my study notes and will be used as a place for triaging new research papers. \n\nI'll use the following icons to differentiate 3D representations:\n* :camera: Multi-view Images\n* :space_invader: Volumetric\n* :game_die: Point Cloud\n* :gem: Polygonal Mesh\n* :pill: Primitive-based\n\nTo find related papers and their relationships, check out [Connected Papers](https:\u002F\u002Fwww.connectedpapers.com\u002F), which provides a neat way to visualize the academic field in a graph representation. \n\n## Get Involved\nTo contribute to this Repo, you may add content through pull requests or open an issue to let me know. \n\n:star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:\u003Cbr>\nWe have also created a Slack workplace for people around the globe to ask questions, share knowledge and facilitate collaborations. Together, I'm sure we can advance this field as a collaborative effort. Join the community with [this link](https:\u002F\u002Fjoin.slack.com\u002Ft\u002F3d-machine-learning\u002Fshared_invite\u002Fzt-4hsgj8zb-G6OKrBcc17QBB9ppYETgCQ).\n\u003Cbr>:star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:\n\n## Table of Contents\n- [Courses](#courses)\n- [Datasets](#datasets)\n  - [3D Models](#3d_models)\n  - [3D Scenes](#3d_scenes)\n- [3D Pose Estimation](#pose_estimation)\n- [Single Object Classification](#single_classification)\n- [Multiple Objects Detection](#multiple_detection)\n- [Scene\u002FObject Semantic Segmentation](#segmentation)\n- [3D Geometry Synthesis\u002FReconstruction](#3d_synthesis)\n  - [Parametric Morphable Model-based methods](#3d_synthesis_model_based)\n  - [Part-based Template Learning methods](#3d_synthesis_template_based)\n  - [Deep Learning Methods](#3d_synthesis_dl_based)\n- [Texture\u002FMaterial Analysis and Synthesis](#material_synthesis)\n- [Style Learning and Transfer](#style_transfer)\n- [Scene Synthesis\u002FReconstruction](#scene_synthesis)\n- [Scene Understanding](#scene_understanding)\n\n\u003Ca name=\"courses\" \u002F>\n\n## Available Courses\n[Stanford CS231A: Computer Vision-From 3D Reconstruction to Recognition (Winter 2018)](http:\u002F\u002Fweb.stanford.edu\u002Fclass\u002Fcs231a\u002F)\n\n[UCSD CSE291-I00: Machine Learning for 3D Data (Winter 2018)](https:\u002F\u002Fcse291-i.github.io\u002Findex.html)\n\n[Stanford CS468: Machine Learning for 3D Data (Spring 2017)](http:\u002F\u002Fgraphics.stanford.edu\u002Fcourses\u002Fcs468-17-spring\u002F)\n\n[MIT 6.838: Shape Analysis (Spring 2017)](http:\u002F\u002Fgroups.csail.mit.edu\u002Fgdpgroup\u002F6838_spring_2017.html)\n\n[Princeton COS 526: Advanced Computer Graphics  (Fall 2010)](https:\u002F\u002Fwww.cs.princeton.edu\u002Fcourses\u002Farchive\u002Ffall10\u002Fcos526\u002Fsyllabus.php)\n\n[Princeton CS597: Geometric Modeling and Analysis (Fall 2003)](https:\u002F\u002Fwww.cs.princeton.edu\u002Fcourses\u002Farchive\u002Ffall03\u002Fcs597D\u002F)\n\n[Geometric Deep Learning](http:\u002F\u002Fgeometricdeeplearning.com\u002F)\n\n[Paper Collection for 3D Understanding](https:\u002F\u002Fwww.cs.princeton.edu\u002Fcourses\u002Farchive\u002Fspring15\u002Fcos598A\u002Fcos598A.html#Estimating)\n\n[CreativeAI: Deep Learning for Graphics](https:\u002F\u002Fgeometry.cs.ucl.ac.uk\u002Fworkshops\u002Fcreativeai\u002F)\n\n\u003Ca name=\"datasets\" \u002F>\n\n## Datasets\nTo see a survey of RGBD datasets, check out Michael Firman's [collection](http:\u002F\u002Fwww.michaelfirman.co.uk\u002FRGBDdatasets\u002Findex.html) as well as the associated paper, [RGBD Datasets: Past, Present and Future](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1604.00999.pdf). Point Cloud Library also has a good dataset [catalogue](https:\u002F\u002Fpointclouds.org\u002F). \n\n\u003Ca name=\"3d_models\" \u002F>\n\n### 3D Models\n\u003Cb>Princeton Shape Benchmark (2003)\u003C\u002Fb> [[Link]](http:\u002F\u002Fshape.cs.princeton.edu\u002Fbenchmark\u002F)\n\u003Cbr>1,814 models collected from the web in .OFF format. Used to evaluating shape-based retrieval and analysis algorithms.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6099efa3b165.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Dataset for IKEA 3D models and aligned images (2013)\u003C\u002Fb> [[Link]](http:\u002F\u002Fikea.csail.mit.edu\u002F)\n\u003Cbr>759 images and 219 models including Sketchup (skp) and Wavefront (obj) files, good for pose estimation.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fikea.csail.mit.edu\u002Fweb_img\u002Fikea_object.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Open Surfaces: A Richly Annotated Catalog of Surface Appearance (SIGGRAPH 2013)\u003C\u002Fb> [[Link]](http:\u002F\u002Fopensurfaces.cs.cornell.edu\u002F)\n\u003Cbr>OpenSurfaces is a large database of annotated surfaces created from real-world consumer photographs. Our annotation framework draws on crowdsourcing to segment surfaces from photos, and then annotate them with rich surface properties, including material, texture and contextual information.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dec5bbf4ae02.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PASCAL3D+ (2014)\u003C\u002Fb> [[Link]](http:\u002F\u002Fcvgl.stanford.edu\u002Fprojects\u002Fpascal3d.html)\n\u003Cbr>12 categories, on average 3k+ objects per category, for 3D object detection and pose estimation.\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e7bc0af1f802.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>ModelNet (2015)\u003C\u002Fb> [[Link]](http:\u002F\u002Fmodelnet.cs.princeton.edu\u002F#)\n\u003Cbr>127915 3D CAD models from 662 categories\n\u003Cbr>ModelNet10: 4899 models from 10 categories\n\u003Cbr>ModelNet40: 12311 models from 40 categories, all are uniformly orientated\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_049b46bd38fa.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>ShapeNet (2015)\u003C\u002Fb> [[Link]](https:\u002F\u002Fwww.shapenet.org\u002F)\n\u003Cbr>3Million+ models and 4K+ categories. A dataset that is large in scale, well organized and richly annotated.\n\u003Cbr>ShapeNetCore [[Link]](http:\u002F\u002Fshapenet.cs.stanford.edu\u002Fshrec16\u002F): 51300 models for 55 categories.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4b4bf06590a6.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>A Large Dataset of Object Scans (2016)\u003C\u002Fb> [[Link]](http:\u002F\u002Fredwood-data.org\u002F3dscan\u002Findex.html)\n\u003Cbr>10K scans in RGBD + reconstructed 3D models in .PLY format.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c6d04b8bb62e.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>ObjectNet3D: A Large Scale Database for 3D Object Recognition (2016)\u003C\u002Fb> [[Link]](http:\u002F\u002Fcvgl.stanford.edu\u002Fprojects\u002Fobjectnet3d\u002F)\n\u003Cbr>100 categories, 90,127 images, 201,888 objects in these images and 44,147 3D shapes. \n\u003Cbr>Tasks: region proposal generation, 2D object detection, joint 2D detection and 3D object pose estimation, and image-based 3D shape retrieval\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_076446608958.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Thingi10K: A Dataset of 10,000 3D-Printing Models (2016)\u003C\u002Fb> [[Link]](https:\u002F\u002Ften-thousand-models.appspot.com\u002F)\n\u003Cbr>10,000 models from featured “things” on thingiverse.com, suitable for testing 3D printing techniques such as structural analysis , shape optimization, or solid geometry operations.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_8c25b834c56a.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>ABC: A Big CAD Model Dataset For Geometric Deep Learning\u003C\u002Fb> [[Link]](https:\u002F\u002Fcs.nyu.edu\u002F~zhongshi\u002Fpublication\u002Fabc-dataset\u002F)[[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.06216)\n\u003Cbr>This work introduce a dataset for geometric deep learning consisting of over 1 million individual (and high quality) geometric models, each associated with accurate ground truth information on the decomposition into patches, explicit sharp feature annotations, and analytic differential properties.\u003Cbr>\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fcs.nyu.edu\u002F~zhongshi\u002Fimg\u002Fabc-dataset.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>ScanObjectNN: A New Benchmark Dataset and Classification Model on Real-World Data (ICCV 2019)\u003C\u002Fb> [[Link]](https:\u002F\u002Fhkust-vgd.github.io\u002Fscanobjectnn\u002F)\n\u003Cbr>\nThis work introduce ScanObjectNN, a new real-world point cloud object dataset based on scanned indoor scene data. The comprehensive benchmark in this work shows that this dataset poses great challenges to existing point cloud classification techniques as objects from real-world scans are often cluttered with background and\u002For are partial due to occlusions. Three key open problems for point cloud object classification are identified, and a new point cloud classification neural network that achieves state-of-the-art performance on classifying objects with cluttered background is proposed.\n\u003Cbr>\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_105eb308140b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>VOCASET: Speech-4D Head Scan Dataset (2019(\u003C\u002Fb> [[Link]](https:\u002F\u002Fvoca.is.tue.mpg.de\u002F)[[Paper]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F510\u002Fpaper_final.pdf)\n\u003Cbr>[VOCASET](https:\u002F\u002Fvoca.is.tue.mpg.de\u002F), is a 4D face dataset with about 29 minutes of 4D scans captured at 60 fps and synchronized audio. The dataset has 12 subjects and 480 sequences of about 3-4 seconds each with sentences chosen from an array of standard protocols that maximize  phonetic  diversity. \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7adaafdb998d.gif\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3D-FUTURE: 3D FUrniture shape with TextURE (2020)\u003C\u002Fb> [[Link]](https:\u002F\u002Ftianchi.aliyun.com\u002Fspecials\u002Fpromotion\u002Falibaba-3d-future?spm=5176.14208320.0.0.66293cf7asRnrR)\n\u003Cbr>[3D-FUTURE](https:\u002F\u002Ftianchi.aliyun.com\u002Fspecials\u002Fpromotion\u002Falibaba-3d-future) contains 20,000+ clean and realistic synthetic scenes in 5,000+ diverse rooms, which include 10,000+ unique high quality 3D instances of furniture with high resolution informative textures developed by professional designers. \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7e2ad2fb60a7.png\" \u002F>\u003C\u002Fp>\n\n\n\u003Cb>Fusion 360 Gallery Dataset (2020)\u003C\u002Fb> [[Link]](https:\u002F\u002Fgithub.com\u002FAutodeskAILab\u002FFusion360GalleryDataset)[[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.02392)\n\u003Cbr>The [Fusion 360 Gallery Dataset](https:\u002F\u002Fgithub.com\u002FAutodeskAILab\u002FFusion360GalleryDataset) contains rich 2D and 3D geometry data derived from parametric CAD models. The Reconstruction Dataset provides sequential construction sequence information from a subset of simple 'sketch and extrude' designs. The Segmentation Dataset provides a segmentation of 3D models based on the CAD modeling operation, including B-Rep format, mesh, and point cloud.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_00d66a3b5af7.jpg\" \u002F>\n\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1aaca440369e.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Mechanical Components Benchmark (2020)\u003C\u002Fb>[[Link]](https:\u002F\u002Fmechanical-components.herokuapp.com)[[Paper]](https:\u002F\u002Fwww.ecva.net\u002Fpapers\u002Feccv_2020\u002Fpapers_ECCV\u002Fpapers\u002F123630171.pdf)\n\u003Cbr>[MCB](https:\u002F\u002Fmechanical-components.herokuapp.com) is a large-scale dataset of 3D objects of mechanical components. It has a total number of 58,696 mechanical components with 68 classes.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fmechanical-components.herokuapp.com\u002Fstatic\u002Fimg\u002Fmain_figure.png\" \u002F>\n\u003C\u002Fp>\n\n\u003Cb>Combinatorial 3D Shape Dataset (2020)\u003C\u002Fb> [[Link]](https:\u002F\u002Fgithub.com\u002FPOSTECH-CVLab\u002FCombinatorial-3D-Shape-Generation)[[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.07414)\n\u003Cbr>[Combinatorial 3D Shape Dataset](https:\u002F\u002Fgithub.com\u002FPOSTECH-CVLab\u002FCombinatorial-3D-Shape-Generation) is composed of 406 instances of 14 classes. Each object in our dataset is considered equivalent to a sequence of primitive placement. Compared to other 3D object datasets, our proposed dataset contains an assembling sequence of unit primitives. It implies that we can quickly obtain a sequential generation process that is a human assembling mechanism. Furthermore, we can sample valid random sequences from a given combinatorial shape after validating the sampled sequences. To sum up, the characteristics of our combinatorial 3D shape dataset are (i) combinatorial, (ii) sequential, (iii) decomposable, and (iv) manipulable.\n\u003Cp align=\"center\">\n\u003Cimg width=\"65%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_41d6146a4782.png\" \u002F>\n\u003C\u002Fp>\n\n\u003Ca name=\"3d_scenes\" \u002F>\n\n### 3D Scenes\n\u003Cb>NYU Depth Dataset V2 (2012)\u003C\u002Fb> [[Link]](https:\u002F\u002Fcs.nyu.edu\u002F~silberman\u002Fdatasets\u002Fnyu_depth_v2.html)\n\u003Cbr>1449 densely labeled pairs of aligned RGB and depth images from Kinect video sequences for a variety of indoor scenes.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fcs.nyu.edu\u002F~silberman\u002Fimages\u002Fnyu_depth_v2_labeled.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SUNRGB-D 3D Object Detection Challenge\u003C\u002Fb> [[Link]](http:\u002F\u002Frgbd.cs.princeton.edu\u002Fchallenge.html)\n\u003Cbr>19 object categories for predicting a 3D bounding box in real world dimension\n\u003Cbr>Training set: 10,355 RGB-D scene images, Testing set: 2860 RGB-D images\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_551157566e91.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SceneNN (2016)\u003C\u002Fb> [[Link]](http:\u002F\u002Fwww.scenenn.net\u002F)\n\u003Cbr>100+ indoor scene meshes with per-vertex and per-pixel annotation.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e8d5ee4a5e62.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>ScanNet (2017)\u003C\u002Fb> [[Link]](http:\u002F\u002Fwww.scan-net.org\u002F)\n\u003Cbr>An RGB-D video dataset containing 2.5 million views in more than 1500 scans, annotated with 3D camera poses, surface reconstructions, and instance-level semantic segmentations.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d4a7035f0e0d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Matterport3D: Learning from RGB-D Data in Indoor Environments (2017)\u003C\u002Fb> [[Link]](https:\u002F\u002Fniessner.github.io\u002FMatterport\u002F)\n\u003Cbr>10,800 panoramic views (in both RGB and depth) from 194,400 RGB-D images of 90 building-scale scenes of private rooms. Instance-level semantic segmentations are provided for region (living room, kitchen) and object (sofa, TV) categories. \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5f4bedf4ad2d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SUNCG: A Large 3D Model Repository for Indoor Scenes (2017)\u003C\u002Fb> [[Link]](http:\u002F\u002Fsuncg.cs.princeton.edu\u002F)\n\u003Cbr>The dataset contains over 45K different scenes with manually created realistic room and furniture layouts. All of the scenes are semantically annotated at the object level.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fsuncg.cs.princeton.edu\u002Ffigures\u002Fdata_full.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>MINOS: Multimodal Indoor Simulator (2017)\u003C\u002Fb> [[Link]](https:\u002F\u002Fgithub.com\u002Fminosworld\u002Fminos)\n\u003Cbr>MINOS is a simulator designed to support the development of multisensory models for goal-directed navigation in complex indoor environments. MINOS leverages large datasets of complex 3D environments and supports flexible configuration of multimodal sensor suites. MINOS supports SUNCG and Matterport3D scenes.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a32adb3a0b39.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Facebook House3D: A Rich and Realistic 3D Environment (2017)\u003C\u002Fb> [[Link]](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FHouse3D)\n\u003Cbr>House3D is a virtual 3D environment which consists of 45K indoor scenes equipped with a diverse set of scene types, layouts and objects sourced from the SUNCG dataset. All 3D objects are fully annotated with category labels. Agents in the environment have access to observations of multiple modalities, including RGB images, depth, segmentation masks and top-down 2D map views.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_72add655ac6b.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>HoME: a Household Multimodal Environment (2017)\u003C\u002Fb> [[Link]](https:\u002F\u002Fhome-platform.github.io\u002F)\n\u003Cbr>HoME integrates over 45,000 diverse 3D house layouts based on the SUNCG dataset, a scale which may facilitate learning, generalization, and transfer. HoME is an open-source, OpenAI Gym-compatible platform extensible to tasks in reinforcement learning, language grounding, sound-based navigation, robotics, multi-agent learning.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fhome-platform.github.io\u002Fassets\u002Foverview.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>AI2-THOR: Photorealistic Interactive Environments for AI Agents\u003C\u002Fb> [[Link]](http:\u002F\u002Fai2thor.allenai.org\u002F)\n\u003Cbr>AI2-THOR is a photo-realistic interactable framework for AI agents. There are a total 120 scenes in version 1.0 of the THOR environment covering four different room categories: kitchens, living rooms, bedrooms, and bathrooms. Each room has a number of actionable objects.\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2176bd000386.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>UnrealCV: Virtual Worlds for Computer Vision (2017)\u003C\u002Fb> [[Link]](http:\u002F\u002Funrealcv.org\u002F)[[Paper]](http:\u002F\u002Fwww.idm.pku.edu.cn\u002Fstaff\u002Fwangyizhou\u002Fpapers\u002FACMMM2017_UnrealCV.pdf)\n\u003Cbr>An open source project to help computer vision researchers build virtual worlds using Unreal Engine 4.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_74ccd6b1ec39.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Gibson Environment: Real-World Perception for Embodied Agents (2018 CVPR) \u003C\u002Fb> [[Link]](http:\u002F\u002Fgibsonenv.stanford.edu\u002F)\n\u003Cbr>This platform provides RGB from 1000 point clouds, as well as multimodal sensor data: surface normal, depth, and for a fraction of the spaces, semantics object annotations. The environment is also RL ready with physics integrated. Using such datasets can further narrow down the discrepency between virtual environment and real world.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d545fe026734.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>InteriorNet: Mega-scale Multi-sensor Photo-realistic Indoor Scenes Dataset\u003C\u002Fb> [[Link]](https:\u002F\u002Finteriornet.org\u002F)\n\u003Cbr>System Overview: an end-to-end pipeline to render an RGB-D-inertial benchmark for large scale interior scene understanding and mapping. Our dataset contains 20M images created by pipeline: (A) We collect around 1 million CAD models provided by world-leading furniture manufacturers. These models have been used in the real-world production. (B) Based on those models, around 1,100 professional designers create around 22 million interior layouts. Most of such layouts have been used in real-world decorations. (C) For each layout, we generate a number of configurations to represent different random lightings and simulation of scene change over time in daily life. (D) We provide an interactive simulator (ViSim) to help for creating ground truth IMU, events, as well as monocular or stereo camera trajectories including hand-drawn, random walking and neural network based realistic trajectory. (E) All supported image sequences and ground truth.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d77814458c07.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Semantic3D\u003C\u002Fb>[[Link]](http:\u002F\u002Fwww.semantic3d.net\u002F)\n\u003Cbr>Large-Scale Point Cloud Classification Benchmark, which provides a large labelled 3D point cloud data set of natural scenes with over 4 billion points in total, and also covers a range of diverse urban scenes.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.semantic3d.net\u002Fimg\u002Ffull_resolution\u002Fsg27_8.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Structured3D: A Large Photo-realistic Dataset for Structured 3D Modeling\u003C\u002Fb> [[Link]](https:\u002F\u002Fstructured3d-dataset.org\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_646ade885892.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3D-FRONT: 3D Furnished Rooms with layOuts and semaNTics\u003C\u002Fb> [[Link]](https:\u002F\u002Ftianchi.aliyun.com\u002Fspecials\u002Fpromotion\u002Falibaba-3d-scene-dataset)\n\u003Cbr>Contains 10,000 houses (or apartments) and ~70,000 rooms with layout information. \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fimg.alicdn.com\u002Ftfs\u002FTB131XOJeL2gK0jSZPhXXahvXXa-2992-2751.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3ThreeDWorld(TDW): A High-Fidelity, Multi-Modal Platform for Interactive Physical Simulation\u003C\u002Fb> [[Link]](http:\u002F\u002Fwww.threedworld.org\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_86724193f8ce.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>MINERVAS: Massive INterior EnviRonments VirtuAl Synthesis\u003C\u002Fb> [[Link]](https:\u002F\u002Fcoohom.github.io\u002FMINERVAS\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ebf8ba38c281.png\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"pose_estimation\" \u002F>\n\n## 3D Pose Estimation\n\u003Cb>Category-Specific Object Reconstruction from a Single Image (2014)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~akar\u002Fcategoryshapes.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7dacad01495f.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Viewpoints and Keypoints (2015)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~shubhtuls\u002Fpapers\u002Fcvpr15vpsKps.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_65905e34f968.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Render for CNN: Viewpoint Estimation in Images Using CNNs Trained with Rendered 3D Model Views (2015 ICCV)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fshapenet.cs.stanford.edu\u002Fprojects\u002FRenderForCNN\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fshapenet.cs.stanford.edu\u002Fprojects\u002FRenderForCNN\u002Fimages\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PoseNet: A Convolutional Network for Real-Time 6-DOF Camera Relocalization (2015)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_iccv_2015\u002Fpapers\u002FKendall_PoseNet_A_Convolutional_ICCV_2015_paper.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fmi.eng.cam.ac.uk\u002Fprojects\u002Frelocalisation\u002Fimages\u002Fmap.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Modeling Uncertainty in Deep Learning for Camera Relocalization (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1509.05909.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_65b961d4ccce.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Robust camera pose estimation by viewpoint classification using deep learning (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Flink.springer.com\u002Farticle\u002F10.1007\u002Fs41095-016-0067-z)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_558ee6146974.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Image-based localization using lstms for structured feature correlation (2017 ICCV)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1611.07890.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6b75e907b4bb.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Image-Based Localization Using Hourglass Networks (2017 ICCV Workshops)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2017_workshops\u002Fpapers\u002Fw17\u002FMelekhov_Image-Based_Localization_Using_ICCV_2017_paper.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1c8285007077.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Geometric loss functions for camera pose regression with deep learning (2017 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1704.00390.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_89e689e051a8.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Generic 3D Representation via Pose Estimation and Matching (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002F3drepresentation.stanford.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_12a2c7cd6774.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3D Bounding Box Estimation Using Deep Learning and Geometry (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1612.00496.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_effa2ed47ba4.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>6-DoF Object Pose from Semantic Keypoints (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.seas.upenn.edu\u002F~pavlakos\u002Fprojects\u002Fobject3d\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.seas.upenn.edu\u002F~pavlakos\u002Fprojects\u002Fobject3d\u002Ffiles\u002Fobject3d-teaser.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Relative Camera Pose Estimation Using Convolutional Neural Networks (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1702.01381.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5a813b8ec775.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002F3dmatch.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dfdd7bea4db4.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Single Image 3D Interpreter Network (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002F3dinterpreter.csail.mit.edu\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fjiajunwu\u002F3dinn)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_26d9f64f28e6.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Multi-view Consistency as Supervisory Signal  for Learning Shape and Pose Prediction (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fshubhtuls.github.io\u002FmvcSnP\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d4f9913e7b0e.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PoseCNN: A Convolutional Neural Network for 6D Object Pose Estimation in Cluttered Scenes (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Frse-lab.cs.washington.edu\u002Fprojects\u002Fposecnn\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fyuxng.github.io\u002FPoseCNN.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Feature Mapping for Learning Fast and Accurate 3D Pose Inference from Synthetic Images (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.03904.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Fencrypted-tbn0.gstatic.com\u002Fimages?q=tbn:ANd9GcTnpyajEhbhrPMc0YpEQzqE8N9E7CW_EVWYA3Bxg46oUEYFf9XvkA\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Pix3D: Dataset and Methods for Single-Image 3D Shape Modeling (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fpix3d.csail.mit.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3a51d7b16939.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3D Pose Estimation and 3D Model Retrieval for Objects in the Wild (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1803.11493.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4bb99a7aef1a.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Deep Object Pose Estimation for Semantic Robotic Grasping of Household Objects (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fresearch.nvidia.com\u002Fpublication\u002F2018-09_Deep-Object-Pose)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5df3b082c4e7.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>MocapNET2: a real-time method that estimates the 3D human pose directly in the popular Bio Vision Hierarchy (BVH) format (2021)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fusers.ics.forth.gr\u002F~argyros\u002Fmypapers\u002F2021_01_ICPR_Qammaz.pdf), [[Code]](https:\u002F\u002Fgithub.com\u002FFORTH-ModelBasedTracker\u002FMocapNET)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_b0eecb3751ea.png\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"single_classification\" \u002F>\n\n## Single Object Classification\n:space_invader: \u003Cb>3D ShapeNets: A Deep Representation for Volumetric Shapes (2015)\u003C\u002Fb> [[Paper]](http:\u002F\u002F3dshapenets.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F3ed23386284a5639cb3e8baaecf496caa766e335\u002F1-Figure1-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>VoxNet: A 3D Convolutional Neural Network for Real-Time Object Recognition (2015)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.dimatura.net\u002Fpublications\u002Fvoxnet_maturana_scherer_iros15.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fdimatura\u002Fvoxnet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cfe207c75926.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Multi-view Convolutional Neural Networks  for 3D Shape Recognition (2015)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Fmvcnn\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fvis-www.cs.umass.edu\u002Fmvcnn\u002Fimages\u002Fmvcnn.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>DeepPano: Deep Panoramic Representation for 3-D Shape Recognition (2015)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fmclab.eic.hust.edu.cn\u002FUpLoadFiles\u002FPapers\u002FDeepPano_SPL2015.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F5a1b5d31905d8cece7b78510f51f3d8bbb063063\u002F1-Figure3-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader::camera: \u003Cb>FusionNet: 3D Object Classification Using Multiple Data Representations (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fstanford.edu\u002F~rezab\u002Fpapers\u002Ffusionnet.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F0aab8fbcef1f0a14f5653d170ca36f4e5aae8010\u002F6-Figure5-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader::camera: \u003Cb>Volumetric and Multi-View CNNs for Object Classification on 3D Data (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1604.03265.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fcharlesq34\u002F3dcnn.torch)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_44abd7dab078.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Generative and Discriminative Voxel Modeling with Convolutional Neural Networks (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1608.04236.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fajbrock\u002FGenerative-and-Discriminative-Voxel-Modeling)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6594154c9914.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Geometric deep learning on graphs and manifolds using mixture model CNNs (2016)\u003C\u002Fb> [[Link]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1611.08402.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fi2.wp.com\u002Fpreferredresearch.jp\u002Fwp-content\u002Fuploads\u002F2017\u002F08\u002Fmonet.png?resize=581%2C155&ssl=1\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3D GAN: Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1610.07584.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fzck119\u002F3dgan-release)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_28bc0d72fbec.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Generative and Discriminative Voxel Modeling with Convolutional Neural Networks (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fgithub.com\u002Fajbrock\u002FGenerative-and-Discriminative-Voxel-Modeling)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5402d3adff7f.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>FPNN: Field Probing Neural Networks for 3D Data (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fyangyanli.github.io\u002FFPNN\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fyangyanli\u002FFPNN)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F15ca7adccf5cd4dc309cdcaa6328f4c429ead337\u002F1-Figure2-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>OctNet: Learning Deep 3D Representations at High Resolutions (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1611.05009.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fgriegler\u002Foctnet)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Fis.tuebingen.mpg.de\u002Fuploads\u002Fpublication\u002Fimage\u002F18921\u002Fimg03.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysis (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwang-ps.github.io\u002FO-CNN) [[Code]](https:\u002F\u002Fgithub.com\u002FMicrosoft\u002FO-CNN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwang-ps.github.io\u002FO-CNN_files\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Orientation-boosted voxel nets for 3D object recognition (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Flmb.informatik.uni-freiburg.de\u002FPublications\u002F2017\u002FSZB17a\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Flmb-freiburg\u002Forion)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_10ac5cfc6bda.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fstanford.edu\u002F~rqi\u002Fpointnet\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fcharlesq34\u002Fpointnet)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4afa19ac3a10.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1706.02413.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fcharlesq34\u002Fpointnet2)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6b557e514b5f.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Feedback Networks (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Ffeedbacknet.stanford.edu\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Famir32002\u002Ffeedback-networks)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a7e4bde24bf3.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Escape from Cells: Deep Kd-Networks for The Recognition of 3D Point Cloud Models (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.arxiv.org\u002Fpdf\u002F1704.01222.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7babf56aef6f.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Dynamic Graph CNN for Learning on Point Clouds (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1801.07829.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_261b1b828b81.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointCNN (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fyangyanli.github.io\u002FPointCNN\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fyangyan.li\u002Fimages\u002Fpaper\u002Fpointcnn.png\" \u002F>\u003C\u002Fp>\n\n:game_die::camera: \u003Cb>A Network Architecture for Point Cloud Classification via Automatic Depth Images Generation (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FRoveri_A_Network_Architecture_CVPR_2018_paper.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_bf3645d0ebf4.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>PointGrid: A Deep Network for 3D Shape Understanding (CVPR 2018) \u003C\u002Fb> [[Paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FLe_PointGrid_A_Deep_CVPR_2018_paper.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Ftrucleduc\u002FPointGrid)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5ae452ff2295.jpeg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb> MeshNet: Mesh Neural Network for 3D Shape Representation (AAAI 2019) \u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1811.11424.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002FYue-Group\u002FMeshNet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cc5a3b00b866.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>SpiderCNN (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fgithub.com\u002Fxyf513\u002FSpiderCNN)[[Code]](https:\u002F\u002Fgithub.com\u002Fxyf513\u002FSpiderCNN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c43c1ffbb08e.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointConv (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fgithub.com\u002FDylanWusee\u002Fpointconv\u002Ftree\u002Fmaster\u002Fimgs)[[Code]](https:\u002F\u002Fgithub.com\u002FDylanWusee\u002Fpointconv\u002Ftree\u002Fmaster\u002Fimgs)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4c58a11c553a.jpeg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>MeshCNN (SIGGRAPH 2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fbit.ly\u002Fmeshcnn)[[Code]](https:\u002F\u002Fgithub.com\u002Franahanocka\u002FMeshCNN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_42897543a21e.gif\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>SampleNet: Differentiable Point Cloud Sampling (CVPR 2020)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2020\u002Fpapers\u002FLang_SampleNet_Differentiable_Point_Cloud_Sampling_CVPR_2020_paper.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fitailang\u002FSampleNet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_b286802523b6.png\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"multiple_detection\" \u002F>\n\n\n## Multiple Objects Detection\n\u003Cb>Sliding Shapes for 3D Object Detection in Depth Images (2014)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fslidingshapes.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9ead403e9d82.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Object Detection in 3D Scenes Using CNNs in Multi-view Images (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fstanford.edu\u002Fclass\u002Fee367\u002FWinter2016\u002FQi_Report.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dc4443a0ddd6.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Deep Sliding Shapes for Amodal 3D Object Detection in RGB-D Images (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fdss.cs.princeton.edu\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fshurans\u002FDeepSlidingShape)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_10dc5d9a18ce.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Three-Dimensional Object Detection and Layout Prediction using Clouds of Oriented Gradients (2016)\u003C\u002Fb> [[CVPR '16 Paper]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2016\u002Fpapers\u002FRen_Three-Dimensional_Object_Detection_CVPR_2016_paper.pdf) [[CVPR '18 Paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FRen_3D_Object_Detection_CVPR_2018_paper.pdf) [[T-PAMI '19 Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1906.04725) \n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d56abb98d169.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>DeepContext: Context-Encoding Neural Pathways  for 3D Holistic Scene Understanding (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fdeepcontext.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_fd0aebf05a84.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SUN RGB-D: A RGB-D Scene Understanding Benchmark Suite (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Frgbd.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_205289780305.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1711.06396.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0b709e99764c.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Frustum PointNets for 3D Object Detection from RGB-D Data (CVPR2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1711.08488.pdf)\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6773cd70533d.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>A^2-Net: Molecular Structure Estimation from Cryo-EM Density Volumes (AAAI2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1901.00785)\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4dd251249b88.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Stereo R-CNN based 3D Object Detection for Autonomous Driving (CVPR2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.09738v1)\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.groundai.com\u002Fmedia\u002Farxiv_projects\u002F515338\u002Fsystem_newnew.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Deep Hough Voting for 3D Object Detection in Point Clouds (ICCV2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1904.09664.pdf) [[code]](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fvotenet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3ad222cc4933.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"segmentation\" \u002F>\n\n## Scene\u002FObject Semantic Segmentation\n\u003Cb>Learning 3D Mesh Segmentation and Labeling (2010)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002FLabelMeshes\u002FLabelMeshes.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F0bf390e2a14f74bcc8838d5fb1c0c4cc60e92eb7\u002F7-Figure7-1.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Unsupervised Co-Segmentation of a Set of Shapes via Descriptor-Space Spectral Clustering (2011)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.cs.sfu.ca\u002F~haoz\u002Fpubs\u002Fsidi_siga11_coseg.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_33f7a41e5e32.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Single-View Reconstruction via Joint Analysis of Image and Shape Collections (2015)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.cs.utexas.edu\u002F~huangqx\u002Fmodeling_sig15.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fhuangqx\u002Fimage_shape_align)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0fd18dbf773d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3D Shape Segmentation with Projective Convolutional Networks (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002Fshapepfcn\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fkalov\u002FShapePFCN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002Fshapepfcn\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Learning Hierarchical Shape Segmentation and Labeling from Online Repositories (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fcs.stanford.edu\u002F~ericyi\u002Fproject_page\u002Fhier_seg\u002Findex.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_17bdea90ac26.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>ScanNet (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1702.04405.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fscannet\u002Fscannet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2bbccdb1f1dc.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fstanford.edu\u002F~rqi\u002Fpointnet\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fcharlesq34\u002Fpointnet)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4afa19ac3a10.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1706.02413.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fcharlesq34\u002Fpointnet2)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6b557e514b5f.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>3D Graph Neural Networks for RGBD Semantic Segmentation (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.cs.toronto.edu\u002F~rjliao\u002Fpapers\u002Ficcv_2017_3DGNN.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"http:\u002F\u002Fwww.fonow.com\u002FImages\u002F2017-10-18\u002F66372-20171018115809740-2125227250.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>3DCNN-DQN-RNN: A Deep Reinforcement Learning Framework for Semantic\nParsing of Large-scale 3D Point Clouds (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1707.06783.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0682e6e111da.png\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>Semantic Segmentation of Indoor Point Clouds using Convolutional Neural Networks (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net\u002FIV-4-W4\u002F101\u002F2017\u002Fisprs-annals-IV-4-W4-101-2017.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"55%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ba0bf1d22d03.png\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>SEGCloud: Semantic Segmentation of 3D Point Clouds (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.07563.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"55%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a8fd8ebc8a1c.png\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>Large-Scale 3D Shape Reconstruction and Segmentation from ShapeNet Core55 (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.06104.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2e57ebd8d643.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Pointwise Convolutional Neural Networks (CVPR 2018)\u003C\u002Fb> [[Link]](http:\u002F\u002Fpointwise.scenenn.net\u002F)\n\u003Cbr>\nWe propose pointwise convolution that performs on-the-fly voxelization for learning local features of a point cloud.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fpointwise.scenenn.net\u002Fimages\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Dynamic Graph CNN for Learning on Point Clouds (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1801.07829.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_261b1b828b81.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointCNN (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fyangyanli.github.io\u002FPointCNN\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fyangyan.li\u002Fimages\u002Fpaper\u002Fpointcnn.png\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader: \u003Cb>3DMV: Joint 3D-Multi-View Prediction for 3D Semantic Scene Segmentation (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1803.10409.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_aaa94f040ad6.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>ScanComplete: Large-Scale Scene Completion and Semantic Segmentation for 3D Scans (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.10215.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c1fbb7324cc1.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die::camera: \u003Cb>SPLATNet: Sparse Lattice Networks for Point Cloud Processing (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1802.08275.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9794752ca75e.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>PointGrid: A Deep Network for 3D Shape Understanding (CVPR 2018) \u003C\u002Fb> [[Paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FLe_PointGrid_A_Deep_CVPR_2018_paper.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Ftrucleduc\u002FPointGrid)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5ae452ff2295.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointConv (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fgithub.com\u002FDylanWusee\u002Fpointconv\u002Ftree\u002Fmaster\u002Fimgs)[[Code]](https:\u002F\u002Fgithub.com\u002FDylanWusee\u002Fpointconv\u002Ftree\u002Fmaster\u002Fimgs)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4c58a11c553a.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>SpiderCNN (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fgithub.com\u002Fxyf513\u002FSpiderCNN)[[Code]](https:\u002F\u002Fgithub.com\u002Fxyf513\u002FSpiderCNN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c43c1ffbb08e.jpeg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3D-SIS: 3D Semantic Instance Segmentation of RGB-D Scans (CVPR 2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1812.07003.pdf)[[Code]](https:\u002F\u002Fgithub.com\u002FSekunde\u002F3D-SIS)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6676ac9721c7.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Real-time Progressive 3D Semantic Segmentation for Indoor Scenes (WACV 2019)\u003C\u002Fb> [[Link]](https:\u002F\u002Fpqhieu.github.io\u002Fresearch\u002Fproseg\u002F)\n\u003Cbr>\nWe propose an efficient yet robust technique for on-the-fly dense reconstruction and semantic segmentation of 3D indoor scenes. Our method is built atop an efficient super-voxel clustering method and a conditional random field with higher-order constraints from structural and object cues, enabling progressive dense semantic segmentation without any precomputation.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpqhieu.github.io\u002Fmedia\u002Fimages\u002Fwacv19\u002Fthumbnail.gif\" \u002F>\u003C\u002Fp>\n\n\n:game_die: \u003Cb>JSIS3D: Joint Semantic-Instance Segmentation of 3D Point Clouds (CVPR 2019)\u003C\u002Fb> [[Link]](https:\u002F\u002Fpqhieu.github.io\u002Fresearch\u002Fjsis3d\u002F)\n\u003Cbr>\nWe jointly address the problems of semantic and instance segmentation of 3D point clouds with a multi-task pointwise network that simultaneously performs two tasks: predicting the semantic classes of 3D points and embedding the points into high-dimensional vectors so that points of the same object instance are represented by similar embeddings. We then propose a multi-value conditional random field model to incorporate the semantic and instance labels and formulate the problem of semantic and instance segmentation as jointly optimising labels in the field model.\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3a72335f6f6e.png\" \u002F>\u003C\u002Fp>\n\n\n:game_die: \u003Cb>ShellNet: Efficient Point Cloud Convolutional Neural Networks using Concentric Shells Statistics (ICCV 2019)\u003C\u002Fb> [[Link]](https:\u002F\u002Fhkust-vgd.github.io\u002Fshellnet\u002F)\n\u003Cbr>\nWe propose an efficient end-to-end permutation invariant convolution for point cloud deep learning. We use statistics from concentric spherical shells to define representative features and resolve the point order ambiguity, allowing traditional convolution to perform efficiently on such features. \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c0164e2d4822.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Rotation Invariant Convolutions for 3D Point Clouds Deep Learning (3DV 2019)\u003C\u002Fb> [[Link]](https:\u002F\u002Fhkust-vgd.github.io\u002Friconv\u002F)\n\u003Cbr>\nWe introduce a novel convolution operator for point clouds that achieves rotation invariance. Our core idea is to use low-level rotation invariant geometric features such as distances and angles to design a convolution operator for point cloud learning. \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_fa95bbb5471c.png\" \u002F>\u003C\u002Fp>\n\n\n\u003Ca name=\"3d_synthesis\" \u002F>\n\n## 3D Model Synthesis\u002FReconstruction\n\n\u003Ca name=\"3d_synthesis_model_based\" \u002F>\n\n### Parametric Morphable Model-based methods\n\n\u003Cb>A Morphable Model For The Synthesis Of 3D Faces (1999)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fgravis.dmi.unibas.ch\u002Fpublications\u002FSigg99\u002Fmorphmod2.pdf)[[Code]](https:\u002F\u002Fgithub.com\u002FMichaelMure\u002F3DMM)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_97fea9364460.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>FLAME: Faces Learned with an Articulated Model and Expressions (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F400\u002Fpaper.pdf)[[Code (Chumpy)]](https:\u002F\u002Fgithub.com\u002FRubikplayer\u002Fflame-fitting)[[Code (TF)]](https:\u002F\u002Fgithub.com\u002FTimoBolkart\u002FTF_FLAME) [[Code (PyTorch)]](https:\u002F\u002Fgithub.com\u002FHavenFeng\u002Fphotometric_optimization)\n\u003Cbr>[FLAME](http:\u002F\u002Fflame.is.tue.mpg.de\u002F) is a lightweight and expressive generic head model learned from over 33,000 of accurately aligned 3D scans. The model combines a linear identity shape space (trained from 3800 scans of human heads) with an articulated neck, jaw, and eyeballs, pose-dependent corrective blendshapes, and additional global expression blendshapes.\nThe code demonstrates how to 1) reconstruct textured 3D faces from images, 2) fit the model to 3D landmarks or registered 3D meshes, or 3) generate 3D face templates for [speech-driven facial animation](https:\u002F\u002Fgithub.com\u002FTimoBolkart\u002Fvoca).\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f7a478b0e3b1.gif\">\u003C\u002Fp>\n\n\u003Cb>The Space of Human Body Shapes: Reconstruction and Parameterization from Range Scans (2003)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fgrail.cs.washington.edu\u002Fprojects\u002Fdigital-human\u002Fpub\u002Fallen03space-submit.pdf)\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F46d39b0e21ae956e4bcb7a789f92be480d45ee12\u002F7-Figure10-1.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SMPL-X: Expressive Body Capture: 3D Hands, Face, and Body from a Single Image (2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F497\u002FSMPL-X.pdf)[[Video]](https:\u002F\u002Fyoutu.be\u002FXyXIEmapWkw)[[Code]](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplify-x)\n\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1f531cecb72a.png\">\u003C\u002Fp>\n\n\u003Cb>PIFuHD: Multi-Level Pixel Aligned Implicit Function for High-Resolution 3D Human Digitization (CVPR 2020)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2004.00452.pdf)[[Video]](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=uEDqCxvF5yc&feature=youtu.be)[[Code]](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fpifuhd)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"\">\u003C\u002Fp>\n\n\n\n\u003Cb>ExPose: Monocular Expressive Body Regression through Body-Driven Attention (2020)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F620\u002F0983.pdf)[[Video]](https:\u002F\u002Fyoutu.be\u002FlNTmHLYTiB8)[[Code]](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fexpose)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_27a06d5bf22e.png\">\u003C\u002Fp>\n\n\u003Cb>Category-Specific Object Reconstruction from a Single Image (2014)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~akar\u002Fcategoryshapes.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e37e34aca9af.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>DeformNet: Free-Form Deformation Network for 3D Shape Reconstruction from a Single Image (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fai.stanford.edu\u002F~haosu\u002Fpapers\u002FSI2PC_arxiv_submit.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cd5e9a27f62f.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Mesh-based Autoencoders for Localized Deformation Component Analysis (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1709.04304.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fqytan.com\u002Fimg\u002Fpoint_conv.jpg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Exploring Generative 3D Shapes Using Autoencoder Networks (Autodesk 2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.autodeskresearch.com\u002Fpublications\u002Fexploring_generative_3d_shapes)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a263365229eb.jpeg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Using Locally Corresponding CAD Models for\nDense 3D Reconstructions from a Single Image (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fci2cv.net\u002Fmedia\u002Fpapers\u002Fchenkong_cvpr_2017.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fchenhsuanlin.bitbucket.io\u002Fimages\u002Frp\u002Fr02.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Compact Model Representation for 3D Reconstruction (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fjhonykaesemodel.com\u002Fpublication\u002F3dv2017\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fjhonykaesemodel.com\u002Fimg\u002Fheaders\u002Foverview.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Image2Mesh: A Learning Framework for Single Image 3D Reconstruction (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1711.10669.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_173d88f1f9b3.jpg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Learning free-form deformations for 3D object reconstruction (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fjhonykaesemodel.com\u002Fpublication\u002Flearning_ffd\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fjhonykaesemodel.com\u002Flearning_ffd_overview.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Variational Autoencoders for Deforming 3D Mesh Models(2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fqytan.com\u002Fpublication\u002Fvae\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fhumanmotion.ict.ac.cn\u002Fpapers\u002F2018P5_VariationalAutoencoders\u002FTeaserImage.jpg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Lions and Tigers and Bears: Capturing Non-Rigid, 3D, Articulated Shape from Images (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Ffiles.is.tue.mpg.de\u002Fblack\u002Fpapers\u002FzuffiCVPR2018.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002F3c1703fe8d.site.internapcdn.net\u002Fnewman\u002Fgfx\u002Fnews\u002Fhires\u002F2018\u002Frealisticava.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"3d_synthesis_template_based\" \u002F>\n\n### Part-based Template Learning methods\n\n\u003Cb>Modeling by Example (2004)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.cs.princeton.edu\u002F~funk\u002Fsig04a.pdf)\n\n\u003Cp align=\"center\">\u003Cimg width=\"20%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3eb31406c2fb.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Model Composition from Interchangeable Components (2007)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.cs.princeton.edu\u002Fcourses\u002Farchive\u002Fspring11\u002Fcos598A\u002Fpdfs\u002FKraevoy07.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d74e9bd41f2f.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Data-Driven Suggestions for Creativity Support in 3D Modeling (2010)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fvladlen.info\u002Fpublications\u002Fdata-driven-suggestions-for-creativity-support-in-3d-modeling\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_26ed1ac6c6cc.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Photo-Inspired Model-Driven 3D Object Modeling (2011)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fkevinkaixu.net\u002Fprojects\u002Fphoto-inspired.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cb1abf33eec7.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Probabilistic Reasoning for Assembly-Based 3D Modeling (2011)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002Fassembly\u002FProbReasoningShapeModeling.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2594d86cc863.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>A Probabilistic Model for Component-Based Shape Synthesis (2012)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002FShapeSynthesis\u002FShapeSynthesis.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1ad8ab461eba.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Structure Recovery by Part Assembly (2012)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fcg.cs.tsinghua.edu.cn\u002FStructureRecovery\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2337c03e1009.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Fit and Diverse: Set Evolution for Inspiring 3D Shape Galleries (2012)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fkevinkaixu.net\u002Fprojects\u002Fcivil.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_64e0ef909e00.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>AttribIt: Content Creation with Semantic Attributes (2013)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002Fattribit\u002FAttribIt.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_03e511a1ca32.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Learning Part-based Templates from Large Collections of 3D Shapes (2013)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fshape.cs.princeton.edu\u002Fvkcorrs\u002Fpapers\u002F13_SIGGRAPH_CorrsTmplt.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9dce4983c214.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Topology-Varying 3D Shape Creation via Structural Blending (2014)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fgruvi.cs.sfu.ca\u002Fproject\u002Ftopo\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_727ef55c2afa.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Estimating Image Depth using Shape Collections (2014)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fvecg.cs.ucl.ac.uk\u002FProjects\u002FSmartGeometry\u002Fimage_shape_net\u002FimageShapeNet_sigg14.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_78d335277e25.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Single-View Reconstruction via Joint Analysis of Image and Shape Collections (2015)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.cs.utexas.edu\u002F~huangqx\u002Fmodeling_sig15.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0fd18dbf773d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Interchangeable Components for Hands-On Assembly Based Modeling (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.cs.umb.edu\u002F~craigyu\u002Fpapers\u002Fhandson_low_res.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0c8ad8d933a5.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Shape Completion from a Single RGBD Image (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.kunzhou.net\u002F2016\u002Fshapecompletion-tvcg16.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dc2165c7f6e2.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"3d_synthesis_dl_based\" \u002F>\n\n### Deep Learning Methods\n\n:camera: \u003Cb>Learning to Generate Chairs, Tables and Cars with Convolutional Networks (2014)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1411.5928.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fzo7.github.io\u002Fimg\u002F2016-09-25-generating-faces\u002Fchairs-model.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Weakly-supervised Disentangling with Recurrent Transformations for 3D View Synthesis (2015, NIPS)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5639-weakly-supervised-disentangling-with-recurrent-transformations-for-3d-view-synthesis.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e19cb6366bcf.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Analysis and synthesis of 3D shape families via deep-learned generative models of surfaces (2015)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~hbhuang\u002Fpublications\u002Fbsm\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpeople.cs.umass.edu\u002F~hbhuang\u002Fpublications\u002Fbsm\u002Fbsm_teaser.jpg\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Weakly-supervised Disentangling with Recurrent Transformations for 3D View Synthesis (2015)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5639-weakly-supervised-disentangling-with-recurrent-transformations-for-3d-view-synthesis.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fjimeiyang\u002FdeepRotator)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F042993c46294a542946c9c1706b7b22deb1d7c43\u002F2-Figure1-1.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Multi-view 3D Models from Single Images with a Convolutional Network (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1511.06702.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Flmb-freiburg\u002Fmv3d)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F3d7ca5ad34f23a5fab16e73e287d1a059dc7ef9a\u002F4-Figure2-1.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>View Synthesis by Appearance Flow (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~tinghuiz\u002Fpapers\u002Feccv16_appflow.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Ftinghuiz\u002Fappearance-flow)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F12280506dc8b5c3ca2db29fc3be694d9a8bef48c\u002F6-Figure2-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Voxlets: Structured Prediction of Unobserved Voxels From a Single Depth Image (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fvisual.cs.ucl.ac.uk\u002Fpubs\u002FdepthPrediction\u002Fhttp:\u002F\u002Fvisual.cs.ucl.ac.uk\u002Fpubs\u002FdepthPrediction\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fmdfirman\u002Fvoxlets)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_167b74476e12.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3D-R2N2: 3D Recurrent Reconstruction Neural Network (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fcvgl.stanford.edu\u002F3d-r2n2\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fchrischoy\u002F3D-R2N2)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_045be81a0041.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Perspective Transformer Nets: Learning Single-View 3D Object Reconstruction without 3D Supervision (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Feng.ucmerced.edu\u002Fpeople\u002Fjyang44\u002Fpapers\u002Fnips16_ptn.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"70%\" src=\"https:\u002F\u002Fsites.google.com\u002Fsite\u002Fskywalkeryxc\u002F_\u002Frsrc\u002F1481104596238\u002Fperspective_transformer_nets\u002Fnetwork_arch.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>TL-Embedding Network: Learning a Predictable and Generative Vector Representation for Objects (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1603.08637.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_12990f95d0f7.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3D GAN: Learning a Probabilistic Latent Space of Object Shapes via 3D Generative-Adversarial Modeling (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1610.07584.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_28bc0d72fbec.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3D Shape Induction from 2D Views of Multiple Objects (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1612.05872.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002Fe78572eeef8b967dec420013c65a6684487c13b2\u002F2-Figure2-1.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Unsupervised Learning of 3D Structure from Images (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1607.00662.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3d31c4c2bd01.jpeg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Generative and Discriminative Voxel Modeling with Convolutional Neural Networks (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1608.04236.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fajbrock\u002FGenerative-and-Discriminative-Voxel-Modeling)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6594154c9914.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Multi-view Supervision for Single-view Reconstruction via Differentiable Ray Consistency (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fshubhtuls.github.io\u002Fdrc\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cf5616b15382.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Synthesizing 3D Shapes via Modeling Multi-View Depth Maps and Silhouettes with Deep Generative Networks (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2017\u002Fpapers\u002FSoltani_Synthesizing_3D_Shapes_CVPR_2017_paper.pdf)  [[Code]](https:\u002F\u002Fgithub.com\u002FAmir-Arsalan\u002FSynthesize3DviaDepthOrSil)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fjiajunwu.com\u002Fimages\u002Fspotlight_3dvae.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Shape Completion using 3D-Encoder-Predictor CNNs and Shape Synthesis (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1612.00101.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fangeladai\u002Fcnncomplete)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ae2512c21f41.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputs (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1703.09438.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Flmb-freiburg\u002Fogn)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F6c2a292bb018a8742cbb0bbc5e23dd0a454ffe3a\u002F2-Figure2-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Hierarchical Surface Prediction for 3D Object Reconstruction (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1704.00710.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fbair.berkeley.edu\u002Fblog\u002Fassets\u002Fhsp\u002Fimage_2.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>OctNetFusion: Learning Depth Fusion from Data (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1704.01047.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fgriegler\u002Foctnetfusion)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2a6ba7734b39.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>A Point Set Generation Network for 3D Object Reconstruction from a Single Image (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fai.stanford.edu\u002F~haosu\u002Fpapers\u002FSI2PC_arxiv_submit.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Ffanhqme\u002FPointSetGeneration)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f7ec0e2bc40b.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Learning Representations and Generative Models for 3D Point Clouds (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1707.02392.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Foptas\u002Flatent_3d_points)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9b47fb6ffd43.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Shape Generation using Spatially Partitioned Point Clouds (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1707.06267.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c122aa1079ce.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PCPNET Learning Local Shape Properties from Raw Point Clouds (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.04954.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4998356c2d5b.jpeg\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Transformation-Grounded Image Generation Network for Novel 3D View Synthesis (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.cs.unc.edu\u002F~eunbyung\u002Ftvsn\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fsilverbottlep\u002Ftvsn)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Feng.ucmerced.edu\u002Fpeople\u002Fjyang44\u002Fpics\u002Fview_synthesis.gif\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Tag Disentangled Generative Adversarial Networks for Object Image Re-rendering (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fstatic.ijcai.org\u002Fproceedings-2017\u002F0404.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3dbc6e9ed119.jpeg\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>3D Shape Reconstruction from Sketches via Multi-view Convolutional Networks (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FSketchModeling\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fhappylun\u002FSketchModeling)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FSketchModeling\u002FSketchModeling_teaser.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Interactive 3D Modeling with a Generative Adversarial Network (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1706.05170.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpbs.twimg.com\u002Fmedia\u002FDCsPKLqXoAEBd-V.jpg\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader: \u003Cb>Weakly supervised 3D Reconstruction with Adversarial Constraint (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1705.10904.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fjgwak\u002FMcRecon)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e55c368f1b75.jpeg\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>SurfNet: Generating 3D shape surfaces using deep residual networks (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1703.04079.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2e4f1df59086.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Learning to Reconstruct Symmetric Shapes using Planar Parameterization of 3D Surface (2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCVW_2019\u002Fpapers\u002FGMDL\u002FJain_Learning_to_Reconstruct_Symmetric_Shapes_using_Planar_Parameterization_of_3D_ICCVW_2019_paper.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fhrdkjain\u002FLearningSymmetricShapes)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7ade75ddca69.png\" \u002F>\u003C\u002Fp>\n\n:pill: \u003Cb>GRASS: Generative Recursive Autoencoders for Shape Structures (SIGGRAPH 2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fkevinkaixu.net\u002Fprojects\u002Fgrass.html) [[Code]](https:\u002F\u002Fgithub.com\u002Fjunli-lj\u002Fgrass) [[code]](https:\u002F\u002Fgithub.com\u002Fkevin-kaixu\u002Fgrass_pytorch)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7ca04019596f.jpg\" \u002F>\u003C\u002Fp>\n\n:pill: \u003Cb> 3D-PRNN: Generating Shape Primitives with Recurrent Neural Networks (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1708.01648.pdf)[[code]](https:\u002F\u002Fgithub.com\u002Fzouchuhang\u002F3D-PRNN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0c52d12cc0ce.jpg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Neural 3D Mesh Renderer (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fhiroharu-kato.com\u002Fprojects_en\u002Fneural_renderer.html) [[Code]](https:\u002F\u002Fgithub.com\u002Fhiroharu-kato\u002Fneural_renderer.git)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7432f17d2162.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>Large-Scale 3D Shape Reconstruction and Segmentation from ShapeNet Core55 (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.06104.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2e57ebd8d643.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Pix2vox: Sketch-Based 3D Exploration with Stacked Generative Adversarial Networks (2017)\u003C\u002Fb> [[Code]](https:\u002F\u002Fgithub.com\u002Fmaxorange\u002Fpix2vox)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_44cdc235354a.gif\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader: \u003Cb>What You Sketch Is What You Get: 3D Sketching using Multi-View Deep Volumetric Prediction (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1707.08390.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Farxiv-sanity-sanity-production.s3.amazonaws.com\u002Frender-output\u002F31631\u002Fx1.png\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader: \u003Cb>MarrNet: 3D Shape Reconstruction via 2.5D Sketches (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fmarrnet.csail.mit.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0a15c76bcd70.jpg\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader::game_die: \u003Cb>Learning a Multi-View Stereo Machine (2017 NIPS)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fbair.berkeley.edu\u002Fblog\u002F2017\u002F09\u002F05\u002Funified-3d\u002F) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cec04b7aff38.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002F3dmatch.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dfdd7bea4db4.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Scaling CNNs for High Resolution Volumetric Reconstruction from a Single Image (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fieeexplore.ieee.org\u002Fdocument\u002F8265323\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_eec17761ca8e.png\" \u002F>\u003C\u002Fp>\n\n:pill: \u003Cb>ComplementMe: Weakly-Supervised Component Suggestions for 3D Modeling (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1708.01841.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1a34f15fedfc.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Learning Descriptor Networks for 3D Shape Synthesis and Analysis (2018 CVPR)\u003C\u002Fb>    [[Project]](http:\u002F\u002Fwww.stat.ucla.edu\u002F~jxie\u002F3DEBM\u002F) [[Paper]](http:\u002F\u002Fwww.stat.ucla.edu\u002F~jxie\u002F3DDescriptorNet\u002F3DDescriptorNet_file\u002Fdoc\u002F3DDescriptorNet.pdf) [[Code](https:\u002F\u002Fgithub.com\u002Fjianwen-xie\u002F3DDescriptorNet)]\n\nAn energy-based 3D shape descriptor network is a deep energy-based model for volumetric shape patterns. The maximum likelihood training of the model follows an “analysis by synthesis” scheme and can be interpreted as a mode seeking and mode shifting process. The model can synthesize 3D shape patterns by sampling from the probability distribution via MCMC such as Langevin dynamics. Experiments demonstrate that the proposed model can generate realistic 3D shape patterns and can be useful for 3D shape analysis.\n\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d703574c7302.png\" \u002F>\u003C\u002Fp> \n\n:game_die: \u003Cb>PU-Net: Point Cloud Upsampling Network (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1801.06761.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fyulequan\u002FPU-Net)\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fappsrv.cse.cuhk.edu.hk\u002F~lqyu\u002Findexpics\u002FPu-Net.png\" \u002F>\u003C\u002Fp> \n\n:camera::space_invader: \u003Cb>Multi-view Consistency as Supervisory Signal  for Learning Shape and Pose Prediction (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fshubhtuls.github.io\u002FmvcSnP\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d4f9913e7b0e.png\" \u002F>\u003C\u002Fp>\n\n:camera::game_die: \u003Cb>Object-Centric Photometric Bundle Adjustment with Deep Shape Prior (2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fci2cv.net\u002Fmedia\u002Fpapers\u002FWACV18.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fchenhsuanlin.bitbucket.io\u002Fimages\u002Frp\u002Fr06.png\" \u002F>\u003C\u002Fp>\n\n:camera::game_die: \u003Cb>Learning Efficient Point Cloud Generation for Dense 3D Object Reconstruction (2018 AAAI)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fchenhsuanlin.bitbucket.io\u002F3D-point-cloud-generation\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fchenhsuanlin.bitbucket.io\u002Fimages\u002Frp\u002Fr05.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fgithub.com\u002Fnywang16\u002FPixel2Mesh)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.groundai.com\u002Fmedia\u002Farxiv_projects\u002F188911\u002Fx2.png.750x0_q75_crop.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>AtlasNet: A Papier-Mâché Approach to Learning 3D Surface Generation (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fimagine.enpc.fr\u002F~groueixt\u002Fatlasnet\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002FThibaultGROUEIX\u002FAtlasNet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_42a5cdeab3e0.png\" \u002F>\u003C\u002Fp>\n\n:space_invader::gem: \u003Cb>Deep Marching Cubes: Learning Explicit Surface Representations (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.cvlibs.net\u002Fpublications\u002FLiao2018CVPR.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a7bbd98df702.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Im2Avatar: Colorful 3D Reconstruction from a Single Image (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.06375v1.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d04e2120aeb8.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Learning Category-Specific Mesh Reconstruction  from Image Collections (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fakanazawa.github.io\u002Fcmr\u002F#)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6cea158eef63.png\" \u002F>\u003C\u002Fp>\n\n:pill: \u003Cb>CSGNet: Neural Shape Parser for Constructive Solid Geometry (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.08290.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_241b86ffd4d1.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Text2Shape: Generating Shapes from Natural Language by Learning Joint Embeddings (2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002Ftext2shape.stanford.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f1147de2f0d0.png\" \u002F>\u003C\u002Fp>\n\n:space_invader::gem::camera: \u003Cb>Multi-View Silhouette and Depth Decomposition for High Resolution 3D Object Representation (2018)\u003C\u002Fb>  [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.09987) [[Code]](https:\u002F\u002Fgithub.com\u002FEdwardSmith1884\u002FMulti-View-Silhouette-and-Depth-Decomposition-for-High-Resolution-3D-Object-Representation)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c2be553de132.png\" \u002F> \u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e158d9ac5d65.png\" \u002F>\u003C\u002Fp>\n\n:space_invader::gem::camera: \u003Cb>Pixels, voxels, and views: A study of shape representations for single view 3D object shape prediction (2018 CVPR)\u003C\u002Fb>  [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.06032)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_37707c78f087.png\" \u002F> \u003C\u002Fp>\n\n:camera::game_die: \u003Cb>Neural scene representation and rendering (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fdeepmind.com\u002Fblog\u002Fneural-scene-representation-and-rendering\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_744614b07eb7.png\" \u002F>\u003C\u002Fp>\n\n:pill: \u003Cb>Im2Struct: Recovering 3D Shape Structure from a Single RGB Image (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.05469.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e220612e0701.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>FoldingNet: Point Cloud Auto-encoder via Deep Grid Deformation (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.07262.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1c68aa17af73.jpg\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader: \u003Cb>Pix3D: Dataset and Methods for Single-Image 3D Shape Modeling (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fpix3d.csail.mit.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9594d245952d.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>3D-RCNN: Instance-level 3D Object Reconstruction via Render-and-Compare (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002FCameraReady\u002F1128.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3911fb42933a.jpeg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Matryoshka Networks: Predicting 3D Geometry via Nested Shape Layers (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.10975.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_48312309cc04.jpeg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>\t\nDeformable Shape Completion with Graph Convolutional Autoencoders (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.00268v1.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_89cec1fa7d46.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Global-to-Local Generative Model for 3D Shapes (SIGGRAPH Asia 2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fvcc.szu.edu.cn\u002Fresearch\u002F2018\u002FG2L)[[Code]](https:\u002F\u002Fgithub.com\u002FHao-HUST\u002FG2LGAN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_53cd78c9e178.jpg\" \u002F>\u003C\u002Fp>\n\n:gem::game_die::space_invader: \u003Cb>ALIGNet: Partial-Shape Agnostic Alignment via Unsupervised Learning (TOG 2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fbit.ly\u002Falignet) [[Code]](https:\u002F\u002Fgithub.com\u002Franahanocka\u002FALIGNet\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5e69e9894f0b.png\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>PointGrid: A Deep Network for 3D Shape Understanding (CVPR 2018) \u003C\u002Fb> [[Paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FLe_PointGrid_A_Deep_CVPR_2018_paper.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Ftrucleduc\u002FPointGrid)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5ae452ff2295.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>GAL: Geometric Adversarial Loss for Single-View 3D-Object Reconstruction (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fxjqi.github.io\u002FGAL.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_aa0abf6416cc.gif\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Visual Object Networks: Image Generation with Disentangled 3D Representation (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F7297-visual-object-networks-image-generation-with-disentangled-3d-representations.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_04cdceb5b42a.jpeg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Learning to Infer and Execute 3D Shape Programs (2019))\u003C\u002Fb> [[Paper]](http:\u002F\u002Fshape2prog.csail.mit.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_068450740104.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Learning to Infer and Execute 3D Shape Programs (2019))\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1901.05103.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f1c282bf915c.jpg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Learning View Priors for Single-view 3D Reconstruction (CVPR 2019)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fhiroharu-kato.com\u002Fprojects_en\u002Fview_prior_learning.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3317f3e24c94.png\" \u002F>\u003C\u002Fp>\n\n:gem::game_die: \u003Cb>Learning Embedding of 3D models with Quadric Loss (BMVC 2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.10250) [[Code]](https:\u002F\u002Fgithub.com\u002Fnitinagarwal\u002FQuadricLoss)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.ics.uci.edu\u002F~agarwal\u002Fbmvc_2019.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>CompoNet: Learning to Generate the Unseen by Part Synthesis and Composition (ICCV 2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.07441)[[Code]](https:\u002F\u002Fgithub.com\u002Fnschor\u002FCompoNet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_474e21cb705e.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>CoMA: Convolutional Mesh Autoencoders (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F439\u002F1285.pdf)[[Code (TF)]](https:\u002F\u002Fgithub.com\u002Fanuragranj\u002Fcoma)[[Code (PyTorch)]](https:\u002F\u002Fgithub.com\u002Fpixelite1201\u002Fpytorch_coma\u002F)[[Code (PyTorch)]](https:\u002F\u002Fgithub.com\u002Fsw-gong\u002Fcoma)\n\u003Cbr>[CoMA](https:\u002F\u002Fcoma.is.tue.mpg.de\u002F) is a versatile model that learns a non-linear representation of a face using spectral convolutions on a mesh surface. CoMA introduces mesh sampling operations that enable a hierarchical mesh representation that captures non-linear variations in shape and expression at multiple scales within the model. \n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Fcoma.is.tue.mpg.de\u002Fuploads\u002Fckeditor\u002Fpictures\u002F91\u002Fcontent_coma_faces.jpg\">\u003C\u002Fp>\n\n\u003Cb>RingNet: 3D Face Reconstruction from Single Images (2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F509\u002Fpaper_camera_ready.pdf)[[Code]](https:\u002F\u002Fgithub.com\u002Fsoubhiksanyal\u002FRingNet)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_179e4f2a621f.gif\">\u003C\u002Fp>\n\n\u003Cb>VOCA: Voice Operated Character Animation (2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F510\u002Fpaper_final.pdf)[[Video]](https:\u002F\u002Fyoutu.be\u002FXceCxf_GyW4)[[Code]](https:\u002F\u002Fgithub.com\u002FTimoBolkart\u002Fvoca)\n\u003Cbr>[VOCA](https:\u002F\u002Fvoca.is.tue.mpg.de\u002F) is a simple and generic speech-driven facial animation framework that works across a range of identities. The codebase demonstrates how to synthesize realistic character animations given an arbitrary speech signal and a static character mesh.\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1cdee30e402b.gif\">\u003C\u002Fp>\n\n:gem: \u003Cb>Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.01210)[[Site]](https:\u002F\u002Fnv-tlabs.github.io\u002FDIB-R\u002F)[[Code]](https:\u002F\u002Fgithub.com\u002Fnv-tlabs\u002FDIB-R)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_56662af983c6.png\"> \u003C\u002Fp>\n\n:gem: \u003Cb>Soft Rasterizer: A Differentiable Renderer for Image-based 3D Reasoning\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.01786)[[Code]](https:\u002F\u002Fgithub.com\u002FShichenLiu\u002FSoftRas)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2b8d73b7f7ee.png\"> \u003C\u002Fp>\n\n\u003Cb>NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis\u003C\u002Fb> [[Project]](http:\u002F\u002Fwww.matthewtancik.com\u002Fnerf)[[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.08934)[[Code]](https:\u002F\u002Fgithub.com\u002Fbmild\u002Fnerf)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ac3328a410fc.png\"> \u003C\u002Fp>\n\n:gem::game_die: \u003Cb>GAMesh: Guided and Augmented Meshing for Deep Point Networks (3DV 2020)\u003C\u002Fb> [[Project]](https:\u002F\u002Fwww.ics.uci.edu\u002F~agarwal\u002FGAMesh\u002F) [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.09774) [[Code]](https:\u002F\u002Fgithub.com\u002Fnitinagarwal\u002FGAMesh)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.ics.uci.edu\u002F~agarwal\u002F3DV_2020.png\" \u002F>\u003C\u002Fp>\n\n\n\n:space_invader: \u003Cb>Generative VoxelNet: Learning Energy-Based Models for 3D Shape Synthesis and Analysis (2020 TPAMI)\u003C\u002Fb>   [[Paper]](http:\u002F\u002Fwww.stat.ucla.edu\u002F~jxie\u002F3DEBM\u002F3DEBM_file\u002Fdoc\u002FgVoxelNet.pdf) \n\nThis paper proposes a deep 3D energy-based model to represent volumetric shapes. The maximum likelihood training of the model follows an “analysis by synthesis” scheme. Experiments demonstrate that the proposed model can generate high-quality 3D shape patterns and can be useful for a wide variety of 3D shape analysis.\n\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ed1b54811bc5.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Generative PointNet: Deep Energy-Based Learning on Unordered Point Sets for 3D Generation, Reconstruction and Classification (2021 CVPR) \u003C\u002Fb> [[Project]](http:\u002F\u002Fwww.stat.ucla.edu\u002F~jxie\u002FGPointNet\u002F) [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2004.01301.pdf) [[Code](https:\u002F\u002Fgithub.com\u002Ffei960922\u002FGPointNet)]\n\nGenerative PointNet is an energy-based model of unordered point clouds, where the energy function is parameterized by an input-permutation-invariant bottom-up neural network. The model can be trained by MCMC-based maximum likelihood learning, or a short-run MCMC toward the energy-based model as a flow-like generator for point cloud reconstruction and interpolation. The learned point cloud representation can be useful for point cloud classification. \n\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_249e4057d91d.png\" \u002F>\u003C\u002Fp>\n\n:game_die: :gem: \u003Cb>Shape My Face: Registering 3D Face Scans by Surface-to-Surface Translation\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.09235) [[Code]](https:\u002F\u002Fgithub.com\u002Fmbahri\u002Fsmf)\n\nShape My Face (SMF) is a point cloud to mesh auto-encoder for the registration of raw human face scans, and the generation of synthetic human faces. SMF leverages a modified PointNet encoder with a visual attention module and differentiable surface sampling to be independent of the original surface representation and reduce the need for pre-processing. Mesh convolution decoders are combined with a specialized PCA model of the mouth, and smoothly blended based on geodesic distances, to create a compact model that is highly robust to noise. SMF is applied to register and perform expression transfer on scans captured in-the-wild with an iPhone depth camera represented either as meshes or point clouds.\n\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_901644743a2c.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>Learning Implicit Fields for Generative Shape Modeling (2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.02822) [[Code]](https:\u002F\u002Fgithub.com\u002Ftimzhang642\u002F3D-Machine-Learning)\n\nWe advocate the use of implicit fields for learning generative models of shapes and introduce an implicit field decoder, called IM-NET, for shape generation, aimed at improving the visual quality of the generated shapes. An implicit field assigns a value to each point in 3D space, so that a shape can be extracted as an iso-surface. IM-NET is trained to perform this assignment by means of a binary classifier. Specifically, it takes a point coordinate, along with a feature vector encoding a shape, and outputs a value which indicates whether the point is outside the shape or not. By replacing conventional decoders by our implicit decoder for representation learning (via IM-AE) and shape generation (via IM-GAN), we demonstrate superior results for tasks such as generative shape modeling, interpolation, and single-view 3D reconstruction, particularly in terms of visual quality.\n\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4d6adb9307d1.png\" \u002F>\u003C\u002Fp>\n\n\n\u003Ca name=\"material_synthesis\" \u002F>\n\n## Texture\u002FMaterial Analysis and Synthesis\n\u003Cb>Texture Synthesis Using Convolutional Neural Networks (2015)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1505.07376.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_af4589e38df6.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Two-Shot SVBRDF Capture for Stationary Materials (SIGGRAPH 2015)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fmediatech.aalto.fi\u002Fpublications\u002Fgraphics\u002FTwoShotSVBRDF\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ac53134c8c13.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Reflectance Modeling by Neural Texture Synthesis (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fmediatech.aalto.fi\u002Fpublications\u002Fgraphics\u002FNeuralSVBRDF\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7d543f74a91b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Modeling Surface Appearance from a Single Photograph using Self-augmented Convolutional Neural Networks (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fmsraig.info\u002F~sanet\u002Fsanet.htm)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fmsraig.info\u002F~sanet\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>High-Resolution Multi-Scale Neural Texture Synthesis (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwxs.ca\u002Fresearch\u002Fmultiscale-neural-synthesis\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_bc1f283c0706.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Reflectance and Natural Illumination from Single Material Specular Objects Using Deep Learning (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fhomes.cs.washington.edu\u002F~krematas\u002FPublications\u002Freflectance-natural-illumination.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.vision.ee.ethz.ch\u002F~georgous\u002Fimages\u002Ftpami17_teaser2.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Joint Material and Illumination Estimation from Photo Sets in the Wild (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.08313.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2063f7bb2d56.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>JWhat Is Around The Camera? (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1611.09325v2.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fhomes.cs.washington.edu\u002F~krematas\u002Fmy_images\u002Farxiv16b_teaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>TextureGAN: Controlling Deep Image Synthesis with Texture Patches (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1706.02823.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Ftexturegan.eye.gatech.edu\u002Fimg\u002Fpaper_figure.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Gaussian Material Synthesis (2018 SIGGRAPH)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fusers.cg.tuwien.ac.at\u002Fzsolnai\u002Fgfx\u002Fgaussian-material-synthesis\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f4018a23c97e.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Non-stationary Texture Synthesis by Adversarial Expansion (2018 SIGGRAPH)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fvcc.szu.edu.cn\u002Fresearch\u002F2018\u002FTexSyn)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_20e5cdcebc0f.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Synthesized Texture Quality Assessment via Multi-scale Spatial and Statistical Texture Attributes of Image and Gradient Magnitude Coefficients (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.08020.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0f2e127a0441.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>LIME: Live Intrinsic Material Estimation (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fgvv.mpi-inf.mpg.de\u002Fprojects\u002FLIME\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fweb.stanford.edu\u002F~zollhoef\u002Fpapers\u002FCVPR18_Material\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Single-Image SVBRDF Capture with a Rendering-Aware Deep Network (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fteam.inria.fr\u002Fgraphdeco\u002Ffr\u002Fprojects\u002Fdeep-materials\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7e4b3b0a498a.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PhotoShape: Photorealistic Materials for Large-Scale Shape Collections (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fkeunhong.com\u002Fpublications\u002Fphotoshape\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fkeunhong.com\u002Fpublications\u002Fphotoshape\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Learning Material-Aware Local Descriptors for 3D Shapes (2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.vovakim.com\u002Fpapers\u002F18_3DV_ShapeMatFeat.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_731bd52f99dd.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>FrankenGAN: Guided Detail Synthesis for Building Mass Models \nusing Style-Synchonized GANs (2018 SIGGRAPH Asia)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fgeometry.cs.ucl.ac.uk\u002Fprojects\u002F2018\u002Ffrankengan\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a4820745e08e.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"style_transfer\" \u002F>\n\n## Style Learning and Transfer\n\u003Cb>Style-Content Separation by Anisotropic Part Scales (2010)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.cs.sfu.ca\u002F~haoz\u002Fpubs\u002Fxu_siga10_style.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ed473ab4310b.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Design Preserving Garment Transfer (2012)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fhal.inria.fr\u002Fhal-00695903\u002Ffile\u002FGarmentTransfer.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2f8a45988aba.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Analogy-Driven 3D Style Transfer (2014)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.chongyangma.com\u002Fpublications\u002Fst\u002Findex.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_adc2b4646158.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Elements of Style: Learning Perceptual Shape Style Similarity (2015)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FStyleSimilarity\u002FStyleSimilarity.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fhappylun\u002FStyleSimilarity)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FStyleSimilarity\u002FStyleSimilarity_teaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Functionality Preserving Shape Style Transfer (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FStyleTransfer\u002FStyleTransfer.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fhappylun\u002FStyleTransfer)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FStyleTransfer\u002FStyleTransfer_teaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Unsupervised Texture Transfer from Images to Model Collections (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fai.stanford.edu\u002F~haosu\u002Fpapers\u002Fsiga16_texture_transfer_small.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9cec00d0dd45.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Learning Detail Transfer based on Geometric Features (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fsurfacedetails.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f92955f96de0.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Co-Locating Style-Defining Elements on 3D Shapes (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fpeople.scs.carleton.ca\u002F~olivervankaick\u002Fpubs\u002Fstyle_elem.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1efe9fed95b9.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Neural 3D Mesh Renderer (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fhiroharu-kato.com\u002Fprojects_en\u002Fneural_renderer.html) [[Code]](https:\u002F\u002Fgithub.com\u002Fhiroharu-kato\u002Fneural_renderer.git)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7432f17d2162.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Appearance Modeling via Proxy-to-Image Alignment (2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fvcc.szu.edu.cn\u002Fresearch\u002F2018\u002FAppMod)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_65b1208a46b1.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images (2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fbigvid.fudan.edu.cn\u002Fpixel2mesh\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5ee4816688b0.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Automatic Unpaired Shape Deformation Transfer (SIGGRAPH Asia 2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fgeometrylearning.com\u002Fausdt\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9ac6e7229911.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3DSNet: Unsupervised Shape-to-Shape 3D Style Transfer (2020)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.13388) [[Code]](https:\u002F\u002Fgithub.com\u002Fethz-asl\u002F3dsnet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_b64e08b146a8.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"scene_synthesis\" \u002F>\n\n## Scene Synthesis\u002FReconstruction\n\u003Cb>Make It Home: Automatic Optimization of Furniture Arrangement (2011, SIGGRAPH)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fpeople.sutd.edu.sg\u002F~saikit\u002Fprojects\u002Ffurniture\u002Findex.html)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7a7e45d17486.gif\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Interactive Furniture Layout Using Interior Design Guidelines (2011)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fgraphics.stanford.edu\u002F~pmerrell\u002FfurnitureLayout.htm)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_072812b3709f.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Synthesizing Open Worlds with Constraints using Locally Annealed Reversible Jump MCMC (2012)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fgraphics.stanford.edu\u002F~lfyg\u002Fowl.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f698fc8e9758.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Example-based Synthesis of 3D Object Arrangements (2012 SIGGRAPH Asia)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fgraphics.stanford.edu\u002Fprojects\u002Fscenesynth\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dd61aa2cc1f0.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Sketch2Scene: Sketch-based Co-retrieval  and Co-placement of 3D Models  (2013)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fsweb.cityu.edu.hk\u002Fhongbofu\u002Fprojects\u002Fsketch2scene_sig13\u002F#.WWWge__ysb0)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_801103499af0.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Action-Driven 3D Indoor Scene Evolution (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.cs.sfu.ca\u002F~haoz\u002Fpubs\u002Fma_siga16_action.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2b82305e6be2.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>The Clutterpalette: An Interactive Tool for Detailing Indoor Scenes (2015)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.cs.umb.edu\u002F~craigyu\u002Fpapers\u002Fclutterpalette.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7352fdc32356.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Image2Scene: Transforming Style of 3D Room (2015)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F2733373.2806274)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e1de6c4e58a1.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Relationship Templates for Creating Scene Variations (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fgeometry.cs.ucl.ac.uk\u002Fprojects\u002F2016\u002Frelationship-templates\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_fdeaa4ef954b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>IM2CAD (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fhomes.cs.washington.edu\u002F~izadinia\u002Fim2cad.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fi.imgur.com\u002FKhtOeuB.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Predicting Complete 3D Models of Indoor Scenes (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1504.02437.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0c429d0e0152.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Complete 3D Scene Parsing from Single RGBD Image (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.09490.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_8d4ca3868c4f.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Raster-to-Vector: Revisiting Floorplan Transformation (2017, ICCV)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.cse.wustl.edu\u002F~chenliu\u002Ffloorplan-transformation.html) [[Code]](https:\u002F\u002Fgithub.com\u002Fart-programmer\u002FFloorplanTransformation)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.cse.wustl.edu\u002F~chenliu\u002Ffloorplan-transformation\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Fully Convolutional Refined Auto-Encoding Generative Adversarial Networks for 3D Multi Object Scenes (2017)\u003C\u002Fb> [[Blog]](https:\u002F\u002Fbecominghuman.ai\u002F3d-multi-object-gan-7b7cee4abf80)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_eeba5f7ae75f.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Adaptive Synthesis of Indoor Scenes via Activity-Associated Object Relation Graphs (2017 SIGGRAPH Asia)\u003C\u002Fb> [[Paper]](http:\u002F\u002Farts.buaa.edu.cn\u002Fprojects\u002Fsa17\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_23f0b545b463.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Automated Interior Design Using a Genetic Algorithm (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpublik.tuwien.ac.at\u002Ffiles\u002Fpublik_262718.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.peterkan.com\u002Fpictures\u002Fteaserq.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SceneSuggest: Context-driven 3D Scene Design (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1703.00061.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_211125dbed0b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>A fully end-to-end deep learning approach for real-time simultaneous 3D reconstruction and material recognition (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1703.04699v1.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7ac82b0fbcfb.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Human-centric Indoor Scene Synthesis Using Stochastic Grammar (2018, CVPR)\u003C\u002Fb>[[Paper]](http:\u002F\u002Fweb.cs.ucla.edu\u002F~syqi\u002Fpublications\u002Fcvpr2018synthesis\u002Fcvpr2018synthesis.pdf) [[Supplementary]](http:\u002F\u002Fweb.cs.ucla.edu\u002F~syqi\u002Fpublications\u002Fcvpr2018synthesis\u002Fcvpr2018synthesis_supplementary.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002FSiyuanQi\u002Fhuman-centric-scene-synthesis)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fweb.cs.ucla.edu\u002F~syqi\u002Fpublications\u002Fthumbnails\u002Fcvpr2018synthesis.gif\" \u002F>\u003C\u002Fp>\n\n:camera::game_die: \u003Cb>FloorNet: A Unified Framework for Floorplan Reconstruction from 3D Scans (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.00090.pdf) [[Code]](http:\u002F\u002Fart-programmer.github.io\u002Ffloornet.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f8e0766d7c5a.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>ScanComplete: Large-Scale Scene Completion and Semantic Segmentation for 3D Scans (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.10215.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_404cfe20d42e.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Deep Convolutional Priors for Indoor Scene Synthesis (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fkwang-ether.github.io\u002Fpdf\u002Fdeepsynth.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5a365413a3a4.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>Fast and Flexible Indoor scene synthesis via Deep Convolutional Generative Models (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1811.12463.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fbrownvc\u002Ffast-synth)\n\u003Cp align=\"center\">\u003Cimg width=\"80%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ed86d51824eb.jpg\" >\u003C\u002Fp>\n\n\u003Cb>Configurable 3D Scene Synthesis and 2D Image Rendering\nwith Per-Pixel Ground Truth using Stochastic Grammars (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1704.00112.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_19fe0afbb156.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Holistic 3D Scene Parsing and Reconstruction from a Single RGB Image (ECCV 2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fsiyuanhuang.com\u002Fholistic_parsing\u002Fmain.html) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fweb.cs.ucla.edu\u002F~syqi\u002Fpublications\u002Fthumbnails\u002Feccv2018scene.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Language-Driven Synthesis of 3D Scenes from Scene Databases (SIGGRAPH Asia 2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.sfu.ca\u002F~agadipat\u002Fpublications\u002F2018\u002FT2S\u002Fproject_page.html) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.sfu.ca\u002F~agadipat\u002Fpublications\u002F2018\u002FT2S\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Deep Generative Modeling for Scene Synthesis via Hybrid Representations (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1808.02084.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_14e57b867c41.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>GRAINS: Generative Recursive Autoencoders for INdoor Scenes (2018)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1807.09193.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.groundai.com\u002Fmedia\u002Farxiv_projects\u002F373503\u002Fnew_pics\u002Fteaserfig.jpg.750x0_q75_crop.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SEETHROUGH: Finding Objects in Heavily Occluded Indoor Scene Images (2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.vovakim.com\u002Fpapers\u002F18_3DVOral_SeeThrough.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4f533697c97d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:space_invader: Scan2CAD: Learning CAD Model Alignment in RGB-D Scans (CVPR 2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1811.11187.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fskanti\u002FScan2CAD)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d0c60afde578.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:gem: Scan2Mesh: From Unstructured Range Scans to 3D Meshes (CVPR 2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1811.10464.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f5ac066de6e4.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:space_invader: 3D-SIC: 3D Semantic Instance Completion for RGB-D Scans (arXiv 2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1904.12012.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.niessnerlab.org\u002Fpapers\u002F2019\u002Fz1sic\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:space_invader: End-to-End CAD Model Retrieval and 9DoF Alignment in 3D Scans (arXiv 2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.04201)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.niessnerlab.org\u002Fpapers\u002F2019\u002Fz2end2end\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>A Survey of 3D Indoor Scene Synthesis (2020) \u003C\u002Fb> [[Paper]](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FShao_Kui_Zhang\u002Fpublication\u002F333135099_A_Survey_of_3D_Indoor_Scene_Synthesis\u002Flinks\u002F5ce13a5492851c4eabad4de0\u002FA-Survey-of-3D-Indoor-Scene-Synthesis.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f80396501ca9.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:pill: :camera: PlanIT: Planning and Instantiating Indoor Scenes with Relation Graph and Spatial Prior Networks (2019) \u003C\u002Fb> [[Paper]](https:\u002F\u002Fkwang-ether.github.io\u002Fpdf\u002Fplanit.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fbrownvc\u002Fplanit)\n\u003Cp align=\"center\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_63f511aed36a.jpg\">\u003C\u002Fp>\n\n\u003Cb>:space_invader: Feature-metric Registration: A Fast Semi-Supervised Approach for Robust Point Cloud Registration without Correspondences (CVPR 2020)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.01014)[[Code]](https:\u002F\u002Fgithub.com\u002FXiaoshuiHuang\u002Ffmr)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fgithub.com\u002FXiaoshuiHuang\u002Fxiaoshuihuang.github.io\u002Fblob\u002Fmaster\u002Fresearch\u002F2020-feature-metric.png?raw=true\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:pill: Human-centric metrics for indoor scene assessment and synthesis (2020) \u003C\u002Fb> [[Paper]](sciencedirect.com\u002Fscience\u002Farticle\u002Fabs\u002Fpii\u002FS1524070320300175)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_48a06950d110.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb> SceneCAD: Predicting Object Alignments and Layouts in RGB-D Scans (2020) \u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2003.12622.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dbdaef7a5b20.jpg\" \u002F>\u003C\u002Fp>\n\n\n\n\n\n\u003Ca name=\"scene_understanding\" \u002F>\n\n## Scene Understanding (Another more detailed [repository](https:\u002F\u002Fgithub.com\u002Fbertjiazheng\u002Fawesome-scene-understanding))\n\n\u003Cb>Recovering the Spatial Layout of Cluttered Rooms (2009)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fdhoiem.cs.illinois.edu\u002Fpublications\u002Ficcv2009_hedau_indoor.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cdacc16c94b7.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Characterizing Structural Relationships in Scenes Using Graph Kernels (2011 SIGGRAPH)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fgraphics.stanford.edu\u002F~mdfisher\u002FgraphKernel.html)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f8389736e53b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Understanding Indoor Scenes Using 3D Geometric Phrases (2013)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fcvgl.stanford.edu\u002Fprojects\u002F3dgp\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a7e0b8ec2241.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Organizing Heterogeneous Scene Collections through Contextual Focal Points (2014 SIGGRAPH)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fkevinkaixu.net\u002Fprojects\u002Ffocal.html)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2977e6dc4db0.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SceneGrok: Inferring Action Maps in 3D Environments (2014, SIGGRAPH)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fgraphics.stanford.edu\u002Fprojects\u002Fscenegrok\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_771bfa73c82d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PanoContext: A Whole-room 3D Context Model for Panoramic Scene Understanding (2014)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fpanocontext.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9e6007d00ea6.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Learning Informative Edge Maps for Indoor Scene Layout Prediction (2015)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fslazebni.cs.illinois.edu\u002Fpublications\u002Ficcv15_informative.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_defcfa4250a0.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Rent3D: Floor-Plan Priors for Monocular Layout Estimation (2015)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.cs.toronto.edu\u002F~fidler\u002Fprojects\u002Frent3D.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_efc4193b0dc7.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>A Coarse-to-Fine Indoor Layout Estimation (CFILE) Method (2016)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpdfs.semanticscholar.org\u002F7024\u002Fa92186b81e6133dc779f497d06877b48d82b.pdf?_ga=2.54181869.497995160.1510977308-665742395.1510465328)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4d3964f8209d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>DeLay: Robust Spatial Layout Estimation for Cluttered Indoor Scenes (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fdeeplayout.stanford.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_14b0542c8f4d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3D Semantic Parsing of Large-Scale Indoor Spaces (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fbuildingparser.stanford.edu\u002Fmethod.html) [[Code]](https:\u002F\u002Fgithub.com\u002Falexsax\u002F2D-3D-Semantics)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fbuildingparser.stanford.edu\u002Fimages\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Single Image 3D Interpreter Network (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002F3dinterpreter.csail.mit.edu\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fjiajunwu\u002F3dinn)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_26d9f64f28e6.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Deep Multi-Modal Image Correspondence Learning (2016)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fwww.cse.wustl.edu\u002F~chenliu\u002Ffloorplan-matching.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ee304c9d0d94.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Physically-Based Rendering for Indoor Scene Understanding Using Convolutional Neural Networks (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002F3dvision.princeton.edu\u002Fprojects\u002F2016\u002FPBRS\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fyindaz\u002Fpbrs) [[Code]](https:\u002F\u002Fgithub.com\u002Fyindaz\u002Fsurface_normal) [[Code]](https:\u002F\u002Fgithub.com\u002Ffyu\u002Fdilation) [[Code]](https:\u002F\u002Fgithub.com\u002Fs9xie\u002Fhed)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e72a7e8c1109.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>RoomNet: End-to-End Room Layout Estimation (2017)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1703.06241.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d40352372be3.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SUN RGB-D: A RGB-D Scene Understanding Benchmark Suite (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Frgbd.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_205289780305.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Semantic Scene Completion from a Single Depth Image (2017)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fsscnet.cs.princeton.edu\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fshurans\u002Fsscnet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_529aad9e6055.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Factoring Shape, Pose, and Layout  from the 2D Image of a 3D Scene (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.01812.pdf) [[Code]](https:\u002F\u002Fshubhtuls.github.io\u002Ffactored3d\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f5019df0ba20.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>LayoutNet: Reconstructing the 3D Room Layout from a Single RGB Image (2018 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1803.08999.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fzouchuhang\u002FLayoutNet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c07b707be767.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PlaneNet: Piece-wise Planar Reconstruction from a Single RGB Image (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fart-programmer.github.io\u002Fplanenet\u002Fpaper.pdf) [[Code]](http:\u002F\u002Fart-programmer.github.io\u002Fplanenet.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4869b4f17c23.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Cross-Domain Self-supervised Multi-task Feature Learning using Synthetic Imagery (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fweb.cs.ucdavis.edu\u002F~yjlee\u002Fprojects\u002Fcvpr2018.pdf) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fjason718.github.io\u002Fproject\u002Fcvpr18\u002Ffiles\u002Fconcept_pic.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Pano2CAD: Room Layout From A Single Panorama Image (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fbjornstenger.github.io\u002Fpapers\u002Fxu_wacv2017.pdf) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.groundai.com\u002Fmedia\u002Farxiv_projects\u002F58924\u002Ffigures\u002FCompare_2b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Automatic 3D Indoor Scene Modeling from Single Panorama (2018 CVPR)\u003C\u002Fb> [[Paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FYang_Automatic_3D_Indoor_CVPR_2018_paper.pdf) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_aad812281fca.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Single-Image Piece-wise Planar 3D Reconstruction via Associative Embedding (2019 CVPR)\u003C\u002Fb> [[Paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1902.09777.pdf) [[Code]](https:\u002F\u002Fgithub.com\u002Fsvip-lab\u002FPlanarReconstruction) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5b7e9eb18828.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3D-Aware Scene Manipulation via Inverse Graphics (NeurIPS 2018)\u003C\u002Fb> [[Paper]](http:\u002F\u002F3dsdn.csail.mit.edu\u002F) [[Code]](https:\u002F\u002Fgithub.com\u002Fsvip-lab\u002FPlanarReconstruction) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_bf98ad17af78.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>3D Scene Reconstruction with Multi-layer Depth and Epipolar Transformers (ICCV 2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fresearch.dshin.org\u002Ficcv19\u002Fmulti-layer-depth\u002F) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ec94229c38af.png\" \u002F>\u003Cbr>\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_15e02a6157b3.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PerspectiveNet: 3D Object Detection from a Single RGB Image via Perspective Points (NIPS 2019)\u003C\u002Fb> [[Paper]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F9093-perspectivenet-3d-object-detection-from-a-single-rgb-image-via-perspective-points.pdf) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fstorage.googleapis.com\u002Fgroundai-web-prod\u002Fmedia\u002Fusers\u002Fuser_288036\u002Fproject_402358\u002Fimages\u002Fx1.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Holistic++ Scene Understanding: Single-view 3D Holistic Scene Parsing and Human Pose Estimation with Human-Object Interaction and Physical Commonsense (ICCV 2019)\u003C\u002Fb> [[Paper & Code]](https:\u002F\u002Fgithub.com\u002Fyixchen\u002Fholistic_scene_human) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4795201634cb.png\" \u002F>\u003C\u002Fp>\n","3D机器学习\n\n近年来，3D机器学习领域取得了巨大进展。该领域是一个融合了计算机视觉、计算机图形学和机器学习的跨学科方向。这个仓库源自我的学习笔记，将用作整理新研究论文的地方。\n\n我将使用以下图标来区分不同的3D表示形式：\n* :camera: 多视角图像\n* :space_invader: 体素表示\n* :game_die: 点云\n* :gem: 多边形网格\n* :pill: 基于基元的表示\n\n要查找相关论文及其相互关系，可以访问[Connected Papers](https:\u002F\u002Fwww.connectedpapers.com\u002F)。它提供了一种以图结构可视化学术领域的便捷方式。\n\n## 参与贡献\n如您希望为本仓库贡献力量，可以通过提交拉取请求添加内容，或直接开一个议题告知我。\n\n:star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:\u003Cbr>\n我们还创建了一个Slack工作区，供全球各地的研究者交流问题、分享知识并促进合作。相信通过共同努力，我们可以携手推动这一领域的进步。请通过[此链接](https:\u002F\u002Fjoin.slack.com\u002Ft\u002F3d-machine-learning\u002Fshared_invite\u002Fzt-4hsgj8zb-G6OKrBcc17QBB9ppYETgCQ)加入社区。\n\u003Cbr>:star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:  :star:\n\n## 目录\n- [课程](#courses)\n- [数据集](#datasets)\n  - [3D模型](#3d_models)\n  - [3D场景](#3d_scenes)\n- [3D姿态估计](#pose_estimation)\n- [单物体分类](#single_classification)\n- [多物体检测](#multiple_detection)\n- [场景\u002F物体语义分割](#segmentation)\n- [3D几何合成\u002F重建](#3d_synthesis)\n  - [基于参数化可变形模型的方法](#3d_synthesis_model_based)\n  - [基于部件模板的学习方法](#3d_synthesis_template_based)\n  - [深度学习方法](#3d_synthesis_dl_based)\n- [纹理\u002F材质分析与合成](#material_synthesis)\n- [风格学习与迁移](#style_transfer)\n- [场景合成\u002F重建](#scene_synthesis)\n- [场景理解](#scene_understanding)\n\n\u003Ca name=\"courses\" \u002F>\n\n## 可用课程\n[斯坦福CS231A：计算机视觉——从3D重建到识别（2018年冬季）](http:\u002F\u002Fweb.stanford.edu\u002Fclass\u002Fcs231a\u002F)\n\n[UCSD CSE291-I00：面向3D数据的机器学习（2018年冬季）](https:\u002F\u002Fcse291-i.github.io\u002Findex.html)\n\n[斯坦福CS468：面向3D数据的机器学习（2017年春季）](http:\u002F\u002Fgraphics.stanford.edu\u002Fcourses\u002Fcs468-17-spring\u002F)\n\n[MIT 6.838：形状分析（2017年春季）](http:\u002F\u002Fgroups.csail.mit.edu\u002Fgdpgroup\u002F6838_spring_2017.html)\n\n[普林斯顿COS 526：高级计算机图形学（2010年秋季）](https:\u002F\u002Fwww.cs.princeton.edu\u002Fcourses\u002Farchive\u002Ffall10\u002Fcos526\u002Fsyllabus.php)\n\n[普林斯顿CS597：几何建模与分析（2003年秋季）](https:\u002F\u002Fwww.cs.princeton.edu\u002Fcourses\u002Farchive\u002Ffall03\u002Fcs597D\u002F)\n\n[几何深度学习](http:\u002F\u002Fgeometricdeeplearning.com\u002F)\n\n[3D理解相关论文合集](https:\u002F\u002Fwww.cs.princeton.edu\u002Fcourses\u002Farchive\u002Fspring15\u002Fcos598A\u002Fcos598A.html#Estimating)\n\n[CreativeAI：面向图形学的深度学习](https:\u002F\u002Fgeometry.cs.ucl.ac.uk\u002Fworkshops\u002Fcreativeai\u002F)\n\n\u003Ca name=\"datasets\" \u002F>\n\n## 数据集\n若想了解RGBD数据集的综述，可以查看Michael Firman的[集合](http:\u002F\u002Fwww.michaelfirman.co.uk\u002FRGBDdatasets\u002Findex.html)，以及相关的论文《RGBD数据集：过去、现在与未来》(https:\u002F\u002Farxiv.org\u002Fpdf\u002F1604.00999.pdf)。Point Cloud Library也有一个不错的数据集[目录](https:\u002F\u002Fpointclouds.org\u002F)。\n\n\u003Ca name=\"3d_models\" \u002F>\n\n### 3D模型\n\u003Cb>普林斯顿形状基准测试集（2003年）\u003C\u002Fb> [[链接]](http:\u002F\u002Fshape.cs.princeton.edu\u002Fbenchmark\u002F)\n\u003Cbr>共收集了1,814个来自网络的.OFF格式模型，用于评估基于形状的检索和分析算法。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6099efa3b165.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>宜家3D模型及配对图像数据集（2013年）\u003C\u002Fb> [[链接]](http:\u002F\u002Fikea.csail.mit.edu\u002F)\n\u003Cbr>包含759张图片和219个模型，文件格式有Sketchup (skp)和Wavefront (obj)，非常适合姿态估计任务。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fikea.csail.mit.edu\u002Fweb_img\u002Fikea_object.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Open Surfaces：表面外观的丰富标注目录（SIGGRAPH 2013）\u003C\u002Fb> [[链接]](http:\u002F\u002Fopensurfaces.cs.cornell.edu\u002F)\n\u003Cbr>OpenSurfaces是一个大型的标注数据库，由真实世界的消费者照片构建而成。其标注框架利用众包技术，从照片中分割出表面，并为其添加丰富的表面属性，包括材质、纹理和上下文信息。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dec5bbf4ae02.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PASCAL3D+（2014年）\u003C\u002Fb> [[链接]](http:\u002F\u002Fcvgl.stanford.edu\u002Fprojects\u002Fpascal3d.html)\n\u003Cbr>涵盖12个类别，平均每类超过3,000个对象，适用于3D目标检测和姿态估计任务。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e7bc0af1f802.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>ModelNet（2015年）\u003C\u002Fb> [[链接]](http:\u002F\u002Fmodelnet.cs.princeton.edu\u002F#)\n\u003Cbr>包含来自662个类别的127,915个3D CAD模型。\n\u003Cbr>ModelNet10：10个类别中的4,899个模型。\n\u003Cbr>ModelNet40：40个类别中的12,311个模型，所有模型均采用统一的方向。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_049b46bd38fa.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>ShapeNet（2015年）\u003C\u002Fb> [[链接]](https:\u002F\u002Fwww.shapenet.org\u002F)\n\u003Cbr>拥有超过300万个模型和4,000多个类别。这是一个规模庞大、组织有序且标注丰富的数据集。\n\u003Cbr>ShapeNetCore [[链接]](http:\u002F\u002Fshapenet.cs.stanford.edu\u002Fshrec16\u002F)：55个类别下的51,300个模型。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4b4bf06590a6.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>大规模物体扫描数据集（2016年）\u003C\u002Fb> [[链接]](http:\u002F\u002Fredwood-data.org\u002F3dscan\u002Findex.html)\n\u003Cbr>包含1万次RGBD扫描，以及以.PLY格式重建的3D模型。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c6d04b8bb62e.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>ObjectNet3D：大规模3D物体识别数据库（2016年）\u003C\u002Fb> [[链接]](http:\u002F\u002Fcvgl.stanford.edu\u002Fprojects\u002Fobjectnet3d\u002F)\n\u003Cbr>涵盖100个类别，90,127张图片，其中包含201,888个物体和44,147个3D形状。\n\u003Cbr>任务包括区域建议生成、2D物体检测、2D检测与3D物体姿态估计的联合任务，以及基于图像的3D形状检索。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_076446608958.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Thingi10K：1万份3D打印模型数据集（2016年）\u003C\u002Fb> [[链接]](https:\u002F\u002Ften-thousand-models.appspot.com\u002F)\n\u003Cbr>包含来自thingiverse.com上精选“事物”的1万份模型，适合用于测试3D打印技术，例如结构分析、形状优化或实体几何操作。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_8c25b834c56a.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>ABC：用于几何深度学习的大型CAD模型数据集\u003C\u002Fb> [[链接]](https:\u002F\u002Fcs.nyu.edu\u002F~zhongshi\u002Fpublication\u002Fabc-dataset\u002F)[[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.06216)\n\u003Cbr>本工作介绍了一个用于几何深度学习的数据集，包含超过100万个独立且高质量的几何模型，每个模型都配有精确的真值信息，包括补丁分解、明确的尖锐特征标注以及解析的微分属性。\u003Cbr>\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fcs.nyu.edu\u002F~zhongshi\u002Fimg\u002Fabc-dataset.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>ScanObjectNN：基于真实世界数据的新基准数据集及分类模型（ICCV 2019）\u003C\u002Fb> [[链接]](https:\u002F\u002Fhkust-vgd.github.io\u002Fscanobjectnn\u002F)\n\u003Cbr>\n本工作介绍了ScanObjectNN，这是一个基于室内场景扫描数据构建的新颖真实世界点云物体数据集。该研究中的全面基准测试表明，由于真实世界的扫描数据中物体常被背景遮挡或因遮挡而呈现不完整状态，这一数据集对现有的点云分类技术提出了巨大挑战。研究还识别出点云物体分类领域的三个关键开放问题，并提出了一种新的点云分类神经网络，该网络在处理具有复杂背景的物体分类任务时达到了最先进的性能。\n\u003Cbr>\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_105eb308140b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>VOCASET：语音-4D头部扫描数据集（2019年）\u003C\u002Fb> [[链接]](https:\u002F\u002Fvoca.is.tue.mpg.de\u002F)[[论文]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F510\u002Fpaper_final.pdf)\n\u003Cbr>[VOCASET](https:\u002F\u002Fvoca.is.tue.mpg.de\u002F)是一个4D面部数据集，包含约29分钟以60帧每秒采集的4D扫描数据，并同步了音频。该数据集共有12名受试者，共480段长约3至4秒的序列，其中的句子选自一系列能够最大化音素多样性的标准协议。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7adaafdb998d.gif\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3D-FUTURE：带有纹理的3D家具形状数据集（2020年）\u003C\u002Fb> [[链接]](https:\u002F\u002Ftianchi.aliyun.com\u002Fspecials\u002Fpromotion\u002Falibaba-3d-future?spm=5176.14208320.0.0.66293cf7asRnrR)\n\u003Cbr>[3D-FUTURE](https:\u002F\u002Ftianchi.aliyun.com\u002Fspecials\u002Fpromotion\u002Falibaba-3d-future)包含5,000多个多样化房间中的20,000多个干净且逼真的合成场景，其中包括由专业设计师开发的10,000多个具有高分辨率、信息丰富纹理的独特高质量3D家具实例。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7e2ad2fb60a7.png\" \u002F>\u003C\u002Fp>\n\n\n\u003Cb>Fusion 360 Gallery 数据集（2020年）\u003C\u002Fb> [[链接]](https:\u002F\u002Fgithub.com\u002FAutodeskAILab\u002FFusion360GalleryDataset)[[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.02392)\n\u003Cbr>[Fusion 360 Gallery 数据集](https:\u002F\u002Fgithub.com\u002FAutodeskAILab\u002FFusion360GalleryDataset)包含了从参数化CAD模型中提取的丰富的2D和3D几何数据。重建数据集提供了一组简单“草图与拉伸”设计的连续构造序列信息。分割数据集则根据CAD建模操作对3D模型进行分割，包括B-Rep格式、网格和点云。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_00d66a3b5af7.jpg\" \u002F>\n\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1aaca440369e.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>机械部件基准数据集（2020年）\u003C\u002Fb>[[链接]](https:\u002F\u002Fmechanical-components.herokuapp.com)[[论文]](https:\u002F\u002Fwww.ecva.net\u002Fpapers\u002Feccv_2020\u002Fpapers_ECCV\u002Fpapers\u002F123630171.pdf)\n\u003Cbr>[MCB](https:\u002F\u002Fmechanical-components.herokuapp.com)是一个大规模的机械部件3D对象数据集。它总共包含58,696个机械部件，分为68个类别。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fmechanical-components.herokuapp.com\u002Fstatic\u002Fimg\u002Fmain_figure.png\" \u002F>\n\u003C\u002Fp>\n\n\u003Cb>组合式3D形状数据集（2020年）\u003C\u002Fb> [[链接]](https:\u002F\u002Fgithub.com\u002FPOSTECH-CVLab\u002FCombinatorial-3D-Shape-Generation)[[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2004.07414)\n\u003Cbr>[组合式3D形状数据集](https:\u002F\u002Fgithub.com\u002FPOSTECH-CVLab\u002FCombinatorial-3D-Shape-Generation)由14个类别的406个实例组成。我们数据集中的每个物体都被视为一系列基本体素放置的等价物。与其他3D物体数据集相比，我们的数据集包含了单元体素的组装序列。这意味着我们可以快速获得一个模拟人类组装过程的顺序生成流程。此外，在验证采样序列的有效性后，我们还可以从给定的组合式形状中随机采样有效的序列。综上所述，我们的组合式3D形状数据集具有以下特点：(i) 组合性，(ii) 顺序性，(iii) 可分解性，以及 (iv) 可操作性。\n\u003Cp align=\"center\">\n\u003Cimg width=\"65%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_41d6146a4782.png\" \u002F>\n\u003C\u002Fp>\n\n\u003Ca name=\"3d_scenes\" \u002F>\n\n\n### 3D场景\n\u003Cb>NYU深度数据集V2（2012年）\u003C\u002Fb> [[链接]](https:\u002F\u002Fcs.nyu.edu\u002F~silberman\u002Fdatasets\u002Fnyu_depth_v2.html)\n\u003Cbr>来自Kinect视频序列的1,449对密集标注的RGB与深度图像配对，涵盖多种室内场景。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fcs.nyu.edu\u002F~silberman\u002Fimages\u002Fnyu_depth_v2_labeled.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SUNRGB-D 3D物体检测挑战赛\u003C\u002Fb> [[链接]](http:\u002F\u002Frgbd.cs.princeton.edu\u002Fchallenge.html)\n\u003Cbr>19个物体类别，用于预测真实世界尺寸的3D边界框\n\u003Cbr>训练集：10,355张RGB-D场景图像，测试集：2,860张RGB-D图像\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_551157566e91.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SceneNN（2016年）\u003C\u002Fb> [[链接]](http:\u002F\u002Fwww.scenenn.net\u002F)\n\u003Cbr>100多个带有顶点级和像素级标注的室内场景网格。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e8d5ee4a5e62.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>ScanNet（2017年）\u003C\u002Fb> [[链接]](http:\u002F\u002Fwww.scan-net.org\u002F)\n\u003Cbr>一个包含250万视角的RGB-D视频数据集，覆盖1,500多次扫描，附有3D相机姿态、表面重建以及实例级语义分割标注。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d4a7035f0e0d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Matterport3D：基于室内环境RGB-D数据的学习（2017年）\u003C\u002Fb> [[链接]](https:\u002F\u002Fniessner.github.io\u002FMatterport\u002F)\n\u003Cbr>来自90个私人房间尺度场景的194,400张RGB-D图像中提取的10,800张全景视图（同时包含RGB和深度信息）。针对区域（客厅、厨房）和物体（沙发、电视）类别提供了实例级语义分割。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5f4bedf4ad2d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SUNCG：用于室内场景的大规模3D模型库（2017年）\u003C\u002Fb> [[链接]](http:\u002F\u002Fsuncg.cs.princeton.edu\u002F)\n\u003Cbr>该数据集包含超过4.5万个不同的场景，每个场景都由人工创建了逼真的房间和家具布局。所有场景都在对象级别进行了语义标注。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fsuncg.cs.princeton.edu\u002Ffigures\u002Fdata_full.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>MINOS：多模态室内模拟器（2017年）\u003C\u002Fb> [[链接]](https:\u002F\u002Fgithub.com\u002Fminosworld\u002Fminos)\n\u003Cbr>MINOS是一款旨在支持在复杂室内环境中开发面向目标的多感官导航模型的模拟器。MINOS利用大规模的复杂3D环境数据集，并支持灵活配置多模态传感器套件。MINOS支持SUNCG和Matterport3D场景。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a32adb3a0b39.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Facebook House3D：丰富且逼真的3D虚拟环境（2017年）\u003C\u002Fb> [[链接]](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002FHouse3D)\n\u003Cbr>House3D是一个虚拟3D环境，包含4.5万个室内场景，涵盖了来自SUNCG数据集的多样化场景类型、布局和物体。所有3D物体均带有完整的类别标签。环境中的智能体可以获取多种模态的观测信息，包括RGB图像、深度图、分割掩码以及俯视二维地图视图。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_72add655ac6b.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>HoME：家庭多模态环境（2017年）\u003C\u002Fb> [[链接]](https:\u002F\u002Fhome-platform.github.io\u002F)\n\u003Cbr>HoME整合了基于SUNCG数据集的超过4.5万个多样化的3D房屋布局，这一规模有助于学习、泛化和迁移。HoME是一个开源、兼容OpenAI Gym的平台，可扩展应用于强化学习、语言接地、基于声音的导航、机器人技术以及多智能体学习等任务。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fhome-platform.github.io\u002Fassets\u002Foverview.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>AI2-THOR：用于AI智能体的写实交互式环境\u003C\u002Fb> [[链接]](http:\u002F\u002Fai2thor.allenai.org\u002F)\n\u003Cbr>AI2-THOR是一个逼真的可交互框架，专为AI智能体设计。THOR环境1.0版本共包含120个场景，覆盖厨房、客厅、卧室和浴室四类房间。每个房间都配备若干可操作对象。\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2176bd000386.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>UnrealCV：用于计算机视觉的虚拟世界（2017年）\u003C\u002Fb> [[链接]](http:\u002F\u002Funrealcv.org\u002F)[[论文]](http:\u002F\u002Fwww.idm.pku.edu.cn\u002Fstaff\u002Fwangyizhou\u002Fpapers\u002FACMMM2017_UnrealCV.pdf)\n\u003Cbr>一个开源项目，旨在帮助计算机视觉研究人员使用虚幻引擎4构建虚拟世界。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_74ccd6b1ec39.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Gibson Environment：具身智能体的真实世界感知环境（2018 CVPR）\u003C\u002Fb> [[链接]](http:\u002F\u002Fgibsonenv.stanford.edu\u002F)\n\u003Cbr>该平台提供来自1000个点云的RGB图像，以及多模态传感器数据：表面法线、深度，并且部分空间还包含语义对象标注。该环境也具备物理引擎，适合强化学习应用。使用此类数据集可以进一步缩小虚拟环境与现实世界之间的差距。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d545fe026734.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>InteriorNet：超大规模多传感器写实室内场景数据集\u003C\u002Fb> [[链接]](https:\u002F\u002Finteriornet.org\u002F)\n\u003Cbr>系统概述：一套端到端的流水线，用于渲染RGB-D惯性基准数据集，以支持大规模室内场景理解和建图。我们的数据集包含2000万张由以下流程生成的图像：(A) 我们收集了约100万个由全球领先的家具制造商提供的CAD模型。这些模型已被实际生产所采用。(B) 基于这些模型，约1100名专业设计师创建了约2200万个室内布局。其中大多数布局已被用于实际装修。(C) 针对每个布局，我们生成了多种配置，以模拟日常生活中不同的随机光照及场景随时间的变化。(D) 我们提供了一个交互式模拟器（ViSim），用于生成IMU、事件以及单目或双目相机轨迹的地面真值，包括手绘轨迹、随机行走轨迹和基于神经网络的逼真轨迹。(E) 所有支持的图像序列及地面真值。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d77814458c07.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Semantic3D\u003C\u002Fb>[[链接]](http:\u002F\u002Fwww.semantic3d.net\u002F)\n\u003Cbr>大规模点云分类基准，提供包含超过40亿个点的自然场景大型标注3D点云数据集，同时也涵盖多种城市场景。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.semantic3d.net\u002Fimg\u002Ffull_resolution\u002Fsg27_8.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Structured3D：用于结构化3D建模的大型写实数据集\u003C\u002Fb> [[链接]](https:\u002F\u002Fstructured3d-dataset.org\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_646ade885892.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3D-FRONT：带布局和语义信息的3D家具化房间\u003C\u002Fb> [[链接]](https:\u002F\u002Ftianchi.aliyun.com\u002Fspecials\u002Fpromotion\u002Falibaba-3d-scene-dataset)\n\u003Cbr>包含1万栋房屋（或公寓）及约7万个带有布局信息的房间。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fimg.alicdn.com\u002Ftfs\u002FTB131XOJeL2gK0jSZPhXXahvXXa-2992-2751.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3ThreeDWorld（TDW）：高保真、多模态的交互式物理仿真平台\u003C\u002Fb> [[链接]](http:\u002F\u002Fwww.threedworld.org\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_86724193f8ce.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>MINERVAS：大规模室内环境虚拟合成\u003C\u002Fb> [[链接]](https:\u002F\u002Fcoohom.github.io\u002FMINERVAS\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ebf8ba38c281.png\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"pose_estimation\" \u002F>\n\n## 3D姿态估计\n\u003Cb>基于单张图像的类别特定物体重建（2014）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~akar\u002Fcategoryshapes.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7dacad01495f.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>视角与关键点（2015）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~shubhtuls\u002Fpapers\u002Fcvpr15vpsKps.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_65905e34f968.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>为CNN而渲染：利用基于渲染的3D模型视图训练的CNN进行图像中的视角估计（2015 ICCV）\u003C\u002Fb> [[论文]](https:\u002F\u002Fshapenet.cs.stanford.edu\u002Fprojects\u002FRenderForCNN\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fshapenet.cs.stanford.edu\u002Fprojects\u002FRenderForCNN\u002Fimages\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PoseNet：用于实时6自由度相机重定位的卷积网络（2015）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_iccv_2015\u002Fpapers\u002FKendall_PoseNet_A_Convolutional_ICCV_2015_paper.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fmi.eng.cam.ac.uk\u002Fprojects\u002Frelocalisation\u002Fimages\u002Fmap.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>深度学习在相机重定位中的不确定性建模（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1509.05909.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_65b961d4ccce.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于深度学习的视角分类实现鲁棒的相机位姿估计（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Flink.springer.com\u002Farticle\u002F10.1007\u002Fs41095-016-0067-z)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_558ee6146974.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于LSTM的结构化特征关联的图像定位（2017 ICCV）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1611.07890.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6b75e907b4bb.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于Hourglass网络的图像定位（2017 ICCV研讨会）\u003C\u002Fb> [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2017_workshops\u002Fpapers\u002Fw17\u002FMelekhov_Image-Based_Localization_Using_ICCV_2017_paper.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1c8285007077.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>深度学习下用于相机位姿回归的几何损失函数（2017 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1704.00390.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_89e689e051a8.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>通过姿态估计与匹配实现通用3D表示（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002F3drepresentation.stanford.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_12a2c7cd6774.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>利用深度学习与几何学进行3D边界框估计（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1612.00496.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_effa2ed47ba4.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于语义关键点的6自由度物体姿态（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.seas.upenn.edu\u002F~pavlakos\u002Fprojects\u002Fobject3d\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.seas.upenn.edu\u002F~pavlakos\u002Fprojects\u002Fobject3d\u002Ffiles\u002Fobject3d-teaser.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>利用卷积神经网络进行相对相机位姿估计（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1702.01381.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5a813b8ec775.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3DMatch：从RGB-D重建中学习局部几何描述子（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002F3dmatch.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dfdd7bea4db4.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>单张图像3D解释网络（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002F3dinterpreter.csail.mit.edu\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fjiajunwu\u002F3dinn)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_26d9f64f28e6.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>多视图一致性作为监督信号用于学习形状和姿态预测（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Fshubhtuls.github.io\u002FmvcSnP\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d4f9913e7b0e.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PoseCNN：用于杂乱场景中6D物体姿态估计的卷积神经网络（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Frse-lab.cs.washington.edu\u002Fprojects\u002Fposecnn\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fyuxng.github.io\u002FPoseCNN.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>用于从合成图像中快速准确推断3D姿态的特征映射（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.03904.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Fencrypted-tbn0.gstatic.com\u002Fimages?q=tbn:ANd9GcTnpyajEhbhrPMc0YpEQzqE8N9E7CW_EVWYA3Bxg46oUEYFf9XvkA\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Pix3D：单张图像3D形状建模的数据集与方法（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fpix3d.csail.mit.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3a51d7b16939.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>野外环境中物体的3D姿态估计与3D模型检索（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1803.11493.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4bb99a7aef1a.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>用于家用物品语义机器人抓取的深度物体姿态估计（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fresearch.nvidia.com\u002Fpublication\u002F2018-09_Deep-Object-Pose)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5df3b082c4e7.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>MocapNET2：一种实时方法，可直接以流行的生物视觉层次结构（BVH）格式估计人体3D姿态（2021）\u003C\u002Fb> [[论文]](http:\u002F\u002Fusers.ics.forth.gr\u002F~argyros\u002Fmypapers\u002F2021_01_ICPR_Qammaz.pdf)，[[代码]](https:\u002F\u002Fgithub.com\u002FFORTH-ModelBasedTracker\u002FMocapNET)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_b0eecb3751ea.png\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"single_classification\" \u002F>\n\n## 单个物体分类\n:space_invader: \u003Cb>3D ShapeNets：用于体素化形状的深度表示（2015）\u003C\u002Fb> [[论文]](http:\u002F\u002F3dshapenets.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F3ed23386284a5639cb3e8baaecf496caa766e335\u002F1-Figure1-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>VoxNet：用于实时物体识别的3D卷积神经网络（2015）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.dimatura.net\u002Fpublications\u002Fvoxnet_maturana_scherer_iros15.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fdimatura\u002Fvoxnet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cfe207c75926.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>用于3D形状识别的多视角卷积神经网络（2015）\u003C\u002Fb> [[论文]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Fmvcnn\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fvis-www.cs.umass.edu\u002Fmvcnn\u002Fimages\u002Fmvcnn.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>DeepPano：用于3D形状识别的深度全景表示（2015）\u003C\u002Fb> [[论文]](http:\u002F\u002Fmclab.eic.hust.edu.cn\u002FUpLoadFiles\u002FPapers\u002FDeepPano_SPL2015.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F5a1b5d31905d8cece7b78510f51f3d8bbb063063\u002F1-Figure3-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader::camera: \u003Cb>FusionNet：利用多种数据表示进行3D物体分类（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Fstanford.edu\u002F~rezab\u002Fpapers\u002Ffusionnet.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F0aab8fbcef1f0a14f5653d170ca36f4e5aae8010\u002F6-Figure5-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader::camera: \u003Cb>用于3D数据物体分类的体素和多视角CNN（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1604.03265.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fcharlesq34\u002F3dcnn.torch)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_44abd7dab078.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>基于卷积神经网络的生成与判别式体素建模（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1608.04236.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fajbrock\u002FGenerative-and-Discriminative-Voxel-Modeling)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6594154c9914.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>使用混合模型CNN在图和流形上进行几何深度学习（2016）\u003C\u002Fb> [[链接]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1611.08402.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fi2.wp.com\u002Fpreferredresearch.jp\u002Fwp-content\u002Fuploads\u002F2017\u002F08\u002Fmonet.png?resize=581%2C155&ssl=1\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3D GAN：通过3D生成对抗建模学习物体形状的概率潜在空间（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1610.07584.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fzck119\u002F3dgan-release)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_28bc0d72fbec.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>基于卷积神经网络的生成与判别式体素建模（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Fgithub.com\u002Fajbrock\u002FGenerative-and-Discriminative-Voxel-Modeling)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5402d3adff7f.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>FPNN：用于3D数据的场探测神经网络（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fyangyanli.github.io\u002FFPNN\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fyangyanli\u002FFPNN)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F15ca7adccf5cd4dc309cdcaa6328f4c429ead337\u002F1-Figure2-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>OctNet：以高分辨率学习深度3D表示（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1611.05009.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fgriegler\u002Foctnet)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Fis.tuebingen.mpg.de\u002Fuploads\u002Fpublication\u002Fimage\u002F18921\u002Fimg03.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>O-CNN：基于八叉树的卷积神经网络用于3D形状分析（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwang-ps.github.io\u002FO-CNN) [[代码]](https:\u002F\u002Fgithub.com\u002FMicrosoft\u002FO-CNN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwang-ps.github.io\u002FO-CNN_files\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>面向3D物体识别的方向增强型体素网络（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Flmb.informatik.uni-freiburg.de\u002FPublications\u002F2017\u002FSZB17a\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Flmb-freiburg\u002Forion)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_10ac5cfc6bda.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointNet：针对3D分类与分割的点云深度学习（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fstanford.edu\u002F~rqi\u002Fpointnet\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fcharlesq34\u002Fpointnet)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4afa19ac3a10.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointNet++：度量空间中点云上的深度层次特征学习（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1706.02413.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fcharlesq34\u002Fpointnet2)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6b557e514b5f.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>反馈网络（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Ffeedbacknet.stanford.edu\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Famir32002\u002Ffeedback-networks)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a7e4bde24bf3.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>逃离细胞：用于3D点云模型识别的深度Kd网络（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.arxiv.org\u002Fpdf\u002F1704.01222.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7babf56aef6f.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>用于点云学习的动态图CNN（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1801.07829.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_261b1b828b81.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointCNN（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fyangyanli.github.io\u002FPointCNN\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fyangyan.li\u002Fimages\u002Fpaper\u002Fpointcnn.png\" \u002F>\u003C\u002Fp>\n\n:game_die::camera: \u003Cb>一种通过自动生成深度图像进行点云分类的网络架构（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FRoveri_A_Network_Architecture_CVPR_2018_paper.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_bf3645d0ebf4.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>PointGrid：用于3D形状理解的深度网络（CVPR 2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FLe_PointGrid_A_Deep_CVPR_2018_paper.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Ftrucleduc\u002FPointGrid)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5ae452ff2295.jpeg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb> MeshNet：用于三维形状表示的网格神经网络（AAAI 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1811.11424.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002FYue-Group\u002FMeshNet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cc5a3b00b866.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>SpiderCNN（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fgithub.com\u002Fxyf513\u002FSpiderCNN)[[代码]](https:\u002F\u002Fgithub.com\u002Fxyf513\u002FSpiderCNN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c43c1ffbb08e.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointConv（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fgithub.com\u002FDylanWusee\u002Fpointconv\u002Ftree\u002Fmaster\u002Fimgs)[[代码]](https:\u002F\u002Fgithub.com\u002FDylanWusee\u002Fpointconv\u002Ftree\u002Fmaster\u002Fimgs)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4c58a11c553a.jpeg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>MeshCNN（SIGGRAPH 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Fbit.ly\u002Fmeshcnn)[[代码]](https:\u002F\u002Fgithub.com\u002Franahanocka\u002FMeshCNN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_42897543a21e.gif\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>SampleNet：可微分点云采样（CVPR 2020）\u003C\u002Fb> [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2020\u002Fpapers\u002FLang_SampleNet_Differentiable_Point_Cloud_Sampling_CVPR_2020_paper.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fitailang\u002FSampleNet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_b286802523b6.png\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"multiple_detection\" \u002F>\n\n\n\n\n## 多物体检测\n\u003Cb>用于深度图像中三维物体检测的滑动形状（2014）\u003C\u002Fb> [[论文]](http:\u002F\u002Fslidingshapes.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9ead403e9d82.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于多视角图像中CNN的三维场景物体检测（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Fstanford.edu\u002Fclass\u002Fee367\u002FWinter2016\u002FQi_Report.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dc4443a0ddd6.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>用于RGB-D图像中遮挡性三维物体检测的深度滑动形状（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fdss.cs.princeton.edu\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fshurans\u002FDeepSlidingShape)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_10dc5d9a18ce.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>使用定向梯度云进行三维物体检测与布局预测（2016）\u003C\u002Fb> [[CVPR '16 论文]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2016\u002Fpapers\u002FRen_Three-Dimensional_Object_Detection_CVPR_2016_paper.pdf) [[CVPR '18 论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FRen_3D_Object_Detection_CVPR_2018_paper.pdf) [[T-PAMI '19 论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1906.04725) \n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d56abb98d169.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>DeepContext：用于三维整体场景理解的上下文编码神经通路（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fdeepcontext.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_fd0aebf05a84.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SUN RGB-D：一个RGB-D场景理解基准套件（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Frgbd.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_205289780305.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>VoxelNet：基于点云的端到端三维物体检测学习（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1711.06396.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0b709e99764c.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>用于RGB-D数据中三维物体检测的Frustum PointNets（CVPR2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1711.08488.pdf)\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6773cd70533d.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>A^2-Net：从冷冻电镜密度体积中估计分子结构（AAAI2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1901.00785)\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4dd251249b88.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于立体R-CNN的自动驾驶三维物体检测（CVPR2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.09738v1)\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.groundai.com\u002Fmedia\u002Farxiv_projects\u002F515338\u002Fsystem_newnew.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>用于点云中三维物体检测的深度霍夫投票法（ICCV2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1904.09664.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fvotenet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3ad222cc4933.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"segmentation\" \u002F>\n\n## 场景\u002F物体语义分割\n\u003Cb>学习三维网格分割与标注（2010）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002FLabelMeshes\u002FLabelMeshes.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F0bf390e2a14f74bcc8838d5fb1c0c4cc60e92eb7\u002F7-Figure7-1.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>通过描述子空间谱聚类对一组形状进行无监督协同分割（2011）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.cs.sfu.ca\u002F~haoz\u002Fpubs\u002Fsidi_siga11_coseg.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_33f7a41e5e32.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>通过图像和形状集合的联合分析进行单视图重建（2015）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.cs.utexas.edu\u002F~huangqx\u002Fmodeling_sig15.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fhuangqx\u002Fimage_shape_align)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0fd18dbf773d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>使用投影卷积网络进行三维形状分割（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002Fshapepfcn\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fkalov\u002FShapePFCN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002Fshapepfcn\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>从在线资源库中学习层次化形状分割与标注（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fcs.stanford.edu\u002F~ericyi\u002Fproject_page\u002Fhier_seg\u002Findex.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_17bdea90ac26.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>ScanNet（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1702.04405.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fscannet\u002Fscannet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2bbccdb1f1dc.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointNet：用于三维分类与分割的点集深度学习（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fstanford.edu\u002F~rqi\u002Fpointnet\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fcharlesq34\u002Fpointnet)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4afa19ac3a10.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointNet++：度量空间中点集的深度层次特征学习（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1706.02413.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fcharlesq34\u002Fpointnet2)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6b557e514b5f.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>用于RGB-D语义分割的3D图神经网络（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.cs.toronto.edu\u002F~rjliao\u002Fpapers\u002Ficcv_2017_3DGNN.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"http:\u002F\u002Fwww.fonow.com\u002FImages\u002F2017-10-18\u002F66372-20171018115809740-2125227250.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>3DCNN-DQN-RNN：大规模3D点云语义解析的深度强化学习框架（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1707.06783.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0682e6e111da.png\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>使用卷积神经网络进行室内点云语义分割（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.isprs-ann-photogramm-remote-sens-spatial-inf-sci.net\u002FIV-4-W4\u002F101\u002F2017\u002Fisprs-annals-IV-4-W4-101-2017.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"55%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ba0bf1d22d03.png\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>SEGCloud：3D点云语义分割（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.07563.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"55%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a8fd8ebc8a1c.png\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>基于ShapeNet Core55的大规模3D形状重建与分割（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.06104.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2e57ebd8d643.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>逐点卷积神经网络（CVPR 2018）\u003C\u002Fb> [[链接]](http:\u002F\u002Fpointwise.scenenn.net\u002F)\n\u003Cbr>\n我们提出了一种逐点卷积方法，该方法可实时进行体素化，以学习点云的局部特征。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fpointwise.scenenn.net\u002Fimages\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>用于点云学习的动态图卷积网络（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1801.07829.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_261b1b828b81.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointCNN（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fyangyanli.github.io\u002FPointCNN\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fyangyan.li\u002Fimages\u002Fpaper\u002Fpointcnn.png\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader: \u003Cb>3DMV：用于3D语义场景分割的联合3D-多视角预测（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1803.10409.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_aaa94f040ad6.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>ScanComplete：3D扫描的大规模场景补全与语义分割（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.10215.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c1fbb7324cc1.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die::camera: \u003Cb>SPLATNet：用于点云处理的稀疏格子网络（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1802.08275.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9794752ca75e.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>PointGrid：用于3D形状理解的深度网络（CVPR 2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FLe_PointGrid_A_Deep_CVPR_2018_paper.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Ftrucleduc\u002FPointGrid)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5ae452ff2295.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PointConv（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fgithub.com\u002FDylanWusee\u002Fpointconv\u002Ftree\u002Fmaster\u002Fimgs)[[代码]](https:\u002F\u002Fgithub.com\u002FDylanWusee\u002Fpointconv\u002Ftree\u002Fmaster\u002Fimgs)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4c58a11c553a.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>SpiderCNN（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fgithub.com\u002Fxyf513\u002FSpiderCNN)[[代码]](https:\u002F\u002Fgithub.com\u002Fxyf513\u002FSpiderCNN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c43c1ffbb08e.jpeg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3D-SIS：RGB-D扫描的3D语义实例分割（CVPR 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1812.07003.pdf)[[代码]](https:\u002F\u002Fgithub.com\u002FSekunde\u002F3D-SIS)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6676ac9721c7.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>室内场景的实时渐进式3D语义分割（WACV 2019）\u003C\u002Fb> [[链接]](https:\u002F\u002Fpqhieu.github.io\u002Fresearch\u002Fproseg\u002F)\n\u003Cbr>\n我们提出了一种高效且鲁棒的技术，用于对3D室内场景进行实时密集重建和语义分割。我们的方法基于高效的超体素聚类算法以及结合结构和物体线索的高阶约束条件的条件随机场，从而能够在无需任何预计算的情况下实现渐进式的密集语义分割。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpqhieu.github.io\u002Fmedia\u002Fimages\u002Fwacv19\u002Fthumbnail.gif\" \u002F>\u003C\u002Fp>\n\n\n:game_die: \u003Cb>JSIS3D：3D点云的联合语义-实例分割（CVPR 2019）\u003C\u002Fb> [[链接]](https:\u002F\u002Fpqhieu.github.io\u002Fresearch\u002Fjsis3d\u002F)\n\u003Cbr>\n我们通过一个多任务逐点网络同时解决3D点云的语义分割和实例分割问题：该网络可以预测3D点的语义类别，并将点嵌入到高维向量中，使得同一对象实例的点由相似的嵌入表示。随后，我们提出了一种多值条件随机场模型来整合语义和实例标签，并将语义和实例分割问题表述为在该场模型中联合优化标签。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3a72335f6f6e.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>ShellNet：基于同心壳统计的高效点云卷积神经网络（ICCV 2019）\u003C\u002Fb> [[链接]](https:\u002F\u002Fhkust-vgd.github.io\u002Fshellnet\u002F)\n\u003Cbr>\n我们提出了一种高效的端到端排列不变卷积方法，用于点云深度学习。通过利用同心球壳的统计信息来定义具有代表性的特征，并解决点顺序的歧义问题，从而使传统的卷积操作能够高效地作用于这些特征。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c0164e2d4822.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>用于3D点云深度学习的旋转不变卷积（3DV 2019）\u003C\u002Fb> [[链接]](https:\u002F\u002Fhkust-vgd.github.io\u002Friconv\u002F)\n\u003Cbr>\n我们提出了一种新颖的点云卷积算子，实现了旋转不变性。我们的核心思想是利用距离和角度等低级的旋转不变几何特征，设计用于点云学习的卷积算子。\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_fa95bbb5471c.png\" \u002F>\u003C\u002Fp>\n\n\n\u003Ca name=\"3d_synthesis\" \u002F>\n\n\n\n## 3D模型合成\u002F重建\n\n\u003Ca name=\"3d_synthesis_model_based\" \u002F>\n\n### 基于参数化可变形模型的方法\n\n\u003Cb>用于3D人脸合成的可变形模型（1999）\u003C\u002Fb> [[论文]](http:\u002F\u002Fgravis.dmi.unibas.ch\u002Fpublications\u002FSigg99\u002Fmorphmod2.pdf)[[代码]](https:\u002F\u002Fgithub.com\u002FMichaelMure\u002F3DMM)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_97fea9364460.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>FLAME：带有关节模型和表情的人脸（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F400\u002Fpaper.pdf)[[代码（Chumpy）]](https:\u002F\u002Fgithub.com\u002FRubikplayer\u002Fflame-fitting)[[代码（TF）]](https:\u002F\u002Fgithub.com\u002FTimoBolkart\u002FTF_FLAME) [[代码（PyTorch）]](https:\u002F\u002Fgithub.com\u002FHavenFeng\u002Fphotometric_optimization)\n\u003Cbr>[FLAME](http:\u002F\u002Fflame.is.tue.mpg.de\u002F) 是一个轻量且富有表现力的通用头部模型，由超过33,000个精确对齐的3D扫描数据训练而成。该模型结合了线性身份形状空间（基于3800个头颅扫描训练）与可关节的颈部、下颌和眼球，以及姿态相关的修正混合形和额外的全局表情混合形。\n代码展示了如何1）从图像中重建带纹理的3D人脸，2）将模型拟合到3D地标或注册好的3D网格上，或者3）生成用于[语音驱动面部动画](https:\u002F\u002Fgithub.com\u002FTimoBolkart\u002Fvoca)的3D人脸模板。\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f7a478b0e3b1.gif\">\u003C\u002Fp>\n\n\u003Cb>人体形态空间：基于范围扫描的重建与参数化（2003）\u003C\u002Fb> [[论文]](http:\u002F\u002Fgrail.cs.washington.edu\u002Fprojects\u002Fdigital-human\u002Fpub\u002Fallen03space-submit.pdf)\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F46d39b0e21ae956e4bcb7a789f92be480d45ee12\u002F7-Figure10-1.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SMPL-X：单张图像中的3D手部、面部和身体表达式捕捉（2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F497\u002FSMPL-X.pdf)[[视频]](https:\u002F\u002Fyoutu.be\u002FXyXIEmapWkw)[[代码]](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplify-x)\n\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1f531cecb72a.png\">\u003C\u002Fp>\n\n\u003Cb>PIFuHD：用于高分辨率人体数字化的多层级像素对齐隐式函数（CVPR 2020）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2004.00452.pdf)[[视频]](https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=uEDqCxvF5yc&feature=youtu.be)[[代码]](https:\u002F\u002Fgithub.com\u002Ffacebookresearch\u002Fpifuhd)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"\">\u003C\u002Fp>\n\n\n\n\u003Cb>ExPose：通过体感注意力进行单目表情身体回归（2020）\u003C\u002Fb> [[论文]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F620\u002F0983.pdf)[[视频]](https:\u002F\u002Fyoutu.be\u002FlNTmHLYTiB8)[[代码]](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fexpose)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_27a06d5bf22e.png\">\u003C\u002Fp>\n\n\u003Cb>基于单张图像的特定类别物体重建（2014）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~akar\u002Fcategoryshapes.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e37e34aca9af.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>DeformNet：用于从单张图像重建3D形状的自由变形网络（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fai.stanford.edu\u002F~haosu\u002Fpapers\u002FSI2PC_arxiv_submit.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cd5e9a27f62f.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>基于网格的自编码器用于局部变形成分分析（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1709.04304.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fqytan.com\u002Fimg\u002Fpoint_conv.jpg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>使用自编码器网络探索生成式3D形状（Autodesk 2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.autodeskresearch.com\u002Fpublications\u002Fexploring_generative_3d_shapes)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a263365229eb.jpeg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>利用局部对应的CAD模型进行单张图像的密集3D重建（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fci2cv.net\u002Fmedia\u002Fpapers\u002Fchenkong_cvpr_2017.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fchenhsuanlin.bitbucket.io\u002Fimages\u002Frp\u002Fr02.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>用于3D重建的紧凑模型表示（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Fjhonykaesemodel.com\u002Fpublication\u002F3dv2017\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fjhonykaesemodel.com\u002Fimg\u002Fheaders\u002Foverview.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Image2Mesh：用于单张图像3D重建的学习框架（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1711.10669.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_173d88f1f9b3.jpg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>学习用于3D物体重建的自由变形（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fjhonykaesemodel.com\u002Fpublication\u002Flearning_ffd\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fjhonykaesemodel.com\u002Flearning_ffd_overview.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>用于变形3D网格模型的变分自编码器（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fqytan.com\u002Fpublication\u002Fvae\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fhumanmotion.ict.ac.cn\u002Fpapers\u002F2018P5_VariationalAutoencoders\u002FTeaserImage.jpg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>狮子、老虎和熊：从图像中捕捉非刚性、3D、可关节的形状（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Ffiles.is.tue.mpg.de\u002Fblack\u002Fpapers\u002FzuffiCVPR2018.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002F3c1703fe8d.site.internapcdn.net\u002Fnewman\u002Fgfx\u002Fnews\u002Fhires\u002F2018\u002Frealisticava.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"3d_synthesis_template_based\" \u002F>\n\n### 基于部件的模板学习方法\n\n\u003Cb>基于示例的建模（2004）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.cs.princeton.edu\u002F~funk\u002Fsig04a.pdf)\n\n\u003Cp align=\"center\">\u003Cimg width=\"20%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3eb31406c2fb.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>由可互换组件组成的模型（2007）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.cs.princeton.edu\u002Fcourses\u002Farchive\u002Fspring11\u002Fcos598A\u002Fpdfs\u002FKraevoy07.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d74e9bd41f2f.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>数据驱动的建议：用于3D建模中创意支持（2010）\u003C\u002Fb> [[论文]](http:\u002F\u002Fvladlen.info\u002Fpublications\u002Fdata-driven-suggestions-for-creativity-support-in-3d-modeling\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_26ed1ac6c6cc.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>受照片启发的模型驱动3D对象建模（2011）\u003C\u002Fb> [[论文]](http:\u002F\u002Fkevinkaixu.net\u002Fprojects\u002Fphoto-inspired.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cb1abf33eec7.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于装配的3D建模中的概率推理（2011）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002Fassembly\u002FProbReasoningShapeModeling.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2594d86cc863.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于组件的形状合成的概率模型（2012）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002FShapeSynthesis\u002FShapeSynthesis.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1ad8ab461eba.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>通过部件装配进行结构恢复（2012）\u003C\u002Fb> [[论文]](http:\u002F\u002Fcg.cs.tsinghua.edu.cn\u002FStructureRecovery\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2337c03e1009.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>契合与多样：用于激发灵感的3D形状图库的集合演化（2012）\u003C\u002Fb> [[论文]](http:\u002F\u002Fkevinkaixu.net\u002Fprojects\u002Fcivil.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_64e0ef909e00.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>AttribIt：使用语义属性进行内容创作（2013）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~kalo\u002Fpapers\u002Fattribit\u002FAttribIt.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_03e511a1ca32.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>从大量3D形状集合中学习基于部件的模板（2013）\u003C\u002Fb> [[论文]](http:\u002F\u002Fshape.cs.princeton.edu\u002Fvkcorrs\u002Fpapers\u002F13_SIGGRAPH_CorrsTmplt.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9dce4983c214.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>通过结构混合实现拓扑可变的3D形状创建（2014）\u003C\u002Fb> [[论文]](http:\u002F\u002Fgruvi.cs.sfu.ca\u002Fproject\u002Ftopo\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_727ef55c2afa.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>利用形状集合估计图像深度（2014）\u003C\u002Fb> [[论文]](http:\u002F\u002Fvecg.cs.ucl.ac.uk\u002FProjects\u002FSmartGeometry\u002Fimage_shape_net\u002FimageShapeNet_sigg14.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_78d335277e25.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>通过联合分析图像和形状集合进行单视图重建（2015）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.cs.utexas.edu\u002F~huangqx\u002Fmodeling_sig15.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0fd18dbf773d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>用于动手装配式建模的可互换组件（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.cs.umb.edu\u002F~craigyu\u002Fpapers\u002Fhandson_low_res.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0c8ad8d933a5.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>从单张RGBD图像完成形状（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.kunzhou.net\u002F2016\u002Fshapecompletion-tvcg16.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dc2165c7f6e2.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"3d_synthesis_dl_based\" \u002F>\n\n### 深度学习方法\n\n:camera: \u003Cb>利用卷积网络学习生成椅子、桌子和汽车（2014）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1411.5928.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fzo7.github.io\u002Fimg\u002F2016-09-25-generating-faces\u002Fchairs-model.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>弱监督解耦与循环变换用于3D视图合成（2015，NIPS）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5639-weakly-supervised-disentangling-with-recurrent-transformations-for-3d-view-synthesis.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e19cb6366bcf.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>通过深度学习的表面生成模型对3D形状族进行分析与合成（2015）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~hbhuang\u002Fpublications\u002Fbsm\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpeople.cs.umass.edu\u002F~hbhuang\u002Fpublications\u002Fbsm\u002Fbsm_teaser.jpg\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>弱监督解耦与循环变换用于3D视图合成（2015）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5639-weakly-supervised-disentangling-with-recurrent-transformations-for-3d-view-synthesis.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fjimeiyang\u002FdeepRotator)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F042993c46294a542946c9c1706b7b22deb1d7c43\u002F2-Figure1-1.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>利用卷积网络从单张图像生成多视图3D模型（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1511.06702.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Flmb-freiburg\u002Fmv3d)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F3d7ca5ad34f23a5fab16e73e287d1a059dc7ef9a\u002F4-Figure2-1.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>基于外观流的视图合成（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpeople.eecs.berkeley.edu\u002F~tinghuiz\u002Fpapers\u002Feccv16_appflow.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Ftinghuiz\u002Fappearance-flow)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F12280506dc8b5c3ca2db29fc3be694d9a8bef48c\u002F6-Figure2-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Voxlets：从单张深度图像预测未观测体素（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fvisual.cs.ucl.ac.uk\u002Fpubs\u002FdepthPrediction\u002Fhttp:\u002F\u002Fvisual.cs.ucl.ac.uk\u002Fpubs\u002FdepthPrediction\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fmdfirman\u002Fvoxlets)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_167b74476e12.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3D-R2N2：3D递归重建神经网络（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fcvgl.stanford.edu\u002F3d-r2n2\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fchrischoy\u002F3D-R2N2)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_045be81a0041.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>透视变换网络：无需3D监督的单视图3D物体重建学习（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Feng.ucmerced.edu\u002Fpeople\u002Fjyang44\u002Fpapers\u002Fnips16_ptn.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"70%\" src=\"https:\u002F\u002Fsites.google.com\u002Fsite\u002Fskywalkeryxc\u002F_\u002Frsrc\u002F1481104596238\u002Fperspective_transformer_nets\u002Fnetwork_arch.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>TL-嵌入网络：为物体学习可预测且生成性的向量表示（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1603.08637.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_12990f95d0f7.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3D GAN：通过3D生成对抗建模学习物体形状的概率潜在空间（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1610.07584.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_28bc0d72fbec.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>从多物体的2D视图中推断3D形状（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1612.05872.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002Fe78572eeef8b967dec420013c65a6684487c13b2\u002F2-Figure2-1.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>从图像中无监督学习3D结构（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1607.00662.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3d31c4c2bd01.jpeg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>基于卷积神经网络的生成与判别体素建模（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1608.04236.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fajbrock\u002FGenerative-and-Discriminative-Voxel-Modeling)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6594154c9914.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>通过可微射线一致性实现单视图重建的多视图监督（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Fshubhtuls.github.io\u002Fdrc\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cf5616b15382.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>利用深度生成网络建模多视角深度图和轮廓来合成3D形状（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2017\u002Fpapers\u002FSoltani_Synthesizing_3D_Shapes_CVPR_2017_paper.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002FAmir-Arsalan\u002FSynthesize3DviaDepthOrSil)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fjiajunwu.com\u002Fimages\u002Fspotlight_3dvae.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>使用3D编码器-预测器CNN和形状合成进行形状补全（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1612.00101.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fangeladai\u002Fcnncomplete)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ae2512c21f41.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>八叉树生成网络：用于高分辨率3D输出的高效卷积架构（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1703.09438.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Flmb-freiburg\u002Fogn)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fai2-s2-public.s3.amazonaws.com\u002Ffigures\u002F2016-11-08\u002F6c2a292bb018a8742cbb0bbc5e23dd0a454ffe3a\u002F2-Figure2-1.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>用于3D物体重建的层次化表面预测（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1704.00710.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fbair.berkeley.edu\u002Fblog\u002Fassets\u002Fhsp\u002Fimage_2.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>OctNetFusion：从数据中学习深度融合（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1704.01047.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fgriegler\u002Foctnetfusion)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2a6ba7734b39.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>用于从单张图像重建3D物体的点集生成网络（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fai.stanford.edu\u002F~haosu\u002Fpapers\u002FSI2PC_arxiv_submit.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Ffanhqme\u002FPointSetGeneration)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f7ec0e2bc40b.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>用于3D点云的表示学习和生成模型（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1707.02392.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Foptas\u002Flatent_3d_points)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9b47fb6ffd43.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>使用空间分区点云进行形状生成（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1707.06267.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c122aa1079ce.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>PCPNET：从原始点云中学习局部形状属性（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.04954.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4998356c2d5b.jpeg\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>基于变换的图像生成网络，用于新颖的3D视图合成（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.cs.unc.edu\u002F~eunbyung\u002Ftvsn\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fsilverbottlep\u002Ftvsn)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Feng.ucmerced.edu\u002Fpeople\u002Fjyang44\u002Fpics\u002Fview_synthesis.gif\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>标签解耦生成对抗网络用于物体图像重渲染（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fstatic.ijcai.org\u002Fproceedings-2017\u002F0404.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3dbc6e9ed119.jpeg\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>通过多视图卷积网络从草图重建3D形状（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FSketchModeling\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fhappylun\u002FSketchModeling)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FSketchModeling\u002FSketchModeling_teaser.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>利用生成对抗网络进行交互式3D建模（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1706.05170.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpbs.twimg.com\u002Fmedia\u002FDCsPKLqXoAEBd-V.jpg\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader: \u003Cb>带有对抗约束的弱监督3D重建（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1705.10904.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fjgwak\u002FMcRecon)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e55c368f1b75.jpeg\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>SurfNet：利用深度残差网络生成三维形状表面（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1703.04079.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2e4f1df59086.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>利用三维曲面的平面参数化学习重建对称形状（2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCVW_2019\u002Fpapers\u002FGMDL\u002FJain_Learning_to_Reconstruct_Symmetric_Shapes_using_Planar_Parameterization_of_3D_ICCVW_2019_paper.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fhrdkjain\u002FLearningSymmetricShapes)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7ade75ddca69.png\" \u002F>\u003C\u002Fp>\n\n:pill: \u003Cb>GRASS：用于形状结构的生成式递归自编码器（SIGGRAPH 2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fkevinkaixu.net\u002Fprojects\u002Fgrass.html) [[代码]](https:\u002F\u002Fgithub.com\u002Fjunli-lj\u002Fgrass) [[代码]](https:\u002F\u002Fgithub.com\u002Fkevin-kaixu\u002Fgrass_pytorch)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7ca04019596f.jpg\" \u002F>\u003C\u002Fp>\n\n:pill: \u003Cb>3D-PRNN：利用循环神经网络生成形状基元（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1708.01648.pdf)[[代码]](https:\u002F\u002Fgithub.com\u002Fzouchuhang\u002F3D-PRNN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0c52d12cc0ce.jpg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>神经网络三维网格渲染器（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fhiroharu-kato.com\u002Fprojects_en\u002Fneural_renderer.html) [[代码]](https:\u002F\u002Fgithub.com\u002Fhiroharu-kato\u002Fneural_renderer.git)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7432f17d2162.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>基于ShapeNet Core55的大规模三维形状重建与分割（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.06104.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2e57ebd8d643.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Pix2vox：基于草图的三维探索，采用堆叠生成对抗网络（2017）\u003C\u002Fb> [[代码]](https:\u002F\u002Fgithub.com\u002Fmaxorange\u002Fpix2vox)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_44cdc235354a.gif\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader: \u003Cb>所绘即所得：利用多视角深度体素预测进行三维草图绘制（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1707.08390.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Farxiv-sanity-sanity-production.s3.amazonaws.com\u002Frender-output\u002F31631\u002Fx1.png\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader: \u003Cb>MarrNet：通过2.5D草图进行三维形状重建（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fmarrnet.csail.mit.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0a15c76bcd70.jpg\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader::game_die: \u003Cb>学习多视图立体视觉机器（2017 NIPS）\u003C\u002Fb> [[论文]](http:\u002F\u002Fbair.berkeley.edu\u002Fblog\u002F2017\u002F09\u002F05\u002Funified-3d\u002F) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cec04b7aff38.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3DMatch：从RGB-D重建中学习局部几何描述符（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002F3dmatch.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dfdd7bea4db4.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>为单张图像的高分辨率体积重建扩展CNN（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Fieeexplore.ieee.org\u002Fdocument\u002F8265323\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_eec17761ca8e.png\" \u002F>\u003C\u002Fp>\n\n:pill: \u003Cb>ComplementMe：用于三维建模的弱监督组件建议（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1708.01841.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1a34f15fedfc.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>学习用于三维形状合成与分析的描述符网络（2018 CVPR）\u003C\u002Fb>    [[项目]](http:\u002F\u002Fwww.stat.ucla.edu\u002F~jxie\u002F3DEBM\u002F) [[论文]](http:\u002F\u002Fwww.stat.ucla.edu\u002F~jxie\u002F3DDescriptorNet\u002F3DDescriptorNet_file\u002Fdoc\u002F3DDescriptorNet.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fjianwen-xie\u002F3DDescriptorNet)\n\n基于能量的三维形状描述符网络是一种用于体素化形状模式的深度能量模型。该模型的最大似然训练遵循“以合成促分析”的方案，可被解释为一种寻找和转移模式的过程。通过诸如朗之万动力学之类的MCMC采样方法，该模型可以从概率分布中生成三维形状模式。实验表明，所提出的模型能够生成逼真的三维形状模式，并可用于三维形状分析。\n\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d703574c7302.png\" \u002F>\u003C\u002Fp> \n\n:game_die: \u003Cb>PU-Net：点云上采样网络（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1801.06761.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fyulequan\u002FPU-Net)\n\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fappsrv.cse.cuhk.edu.hk\u002F~lqyu\u002Findexpics\u002FPu-Net.png\" \u002F>\u003C\u002Fp> \n\n:camera::space_invader: \u003Cb>多视图一致性作为监督信号，用于学习形状和姿态预测（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Fshubhtuls.github.io\u002FmvcSnP\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d4f9913e7b0e.png\" \u002F>\u003C\u002Fp>\n\n:camera::game_die: \u003Cb>以对象为中心的光度束调整，结合深度形状先验（2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fci2cv.net\u002Fmedia\u002Fpapers\u002FWACV18.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fchenhsuanlin.bitbucket.io\u002Fimages\u002Frp\u002Fr06.png\" \u002F>\u003C\u002Fp>\n\n:camera::game_die: \u003Cb>学习高效的点云生成技术，用于密集型三维物体重建（2018 AAAI）\u003C\u002Fb> [[论文]](https:\u002F\u002Fchenhsuanlin.bitbucket.io\u002F3D-point-cloud-generation\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fchenhsuanlin.bitbucket.io\u002Fimages\u002Frp\u002Fr05.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Pixel2Mesh：从单张RGB图像生成三维网格模型（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fgithub.com\u002Fnywang16\u002FPixel2Mesh)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.groundai.com\u002Fmedia\u002Farxiv_projects\u002F188911\u002Fx2.png.750x0_q75_crop.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>AtlasNet：一种基于纸浆工艺的学习三维表面生成方法（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fimagine.enpc.fr\u002F~groueixt\u002Fatlasnet\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002FThibaultGROUEIX\u002FAtlasNet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_42a5cdeab3e0.png\" \u002F>\u003C\u002Fp>\n\n:space_invader::gem: \u003Cb>深度Marching Cubes：学习显式表面表示（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.cvlibs.net\u002Fpublications\u002FLiao2018CVPR.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a7bbd98df702.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Im2Avatar：基于单张图像的彩色3D重建（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.06375v1.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d04e2120aeb8.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>从图像集合中学习特定类别的网格重建（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fakanazawa.github.io\u002Fcmr\u002F#)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_6cea158eef63.png\" \u002F>\u003C\u002Fp>\n\n:pill: \u003Cb>CSGNet：用于构造实体几何的神经形状解析器（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.08290.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_241b86ffd4d1.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>Text2Shape：通过学习联合嵌入从自然语言生成形状（2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Ftext2shape.stanford.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f1147de2f0d0.png\" \u002F>\u003C\u002Fp>\n\n:space_invader::gem::camera: \u003Cb>高分辨率3D物体表示的多视角轮廓与深度分解（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1802.09987) [[代码]](https:\u002F\u002Fgithub.com\u002FEdwardSmith1884\u002FMulti-View-Silhouette-and-Depth-Decomposition-for-High-Resolution-3D-Object-Representation)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c2be553de132.png\" \u002F> \u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e158d9ac5d65.png\" \u002F>\u003C\u002Fp>\n\n:space_invader::gem::camera: \u003Cb>像素、体素和视图：单视角3D物体形状预测的形状表示研究（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.06032)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_37707c78f087.png\" \u002F> \u003C\u002Fp>\n\n:camera::game_die: \u003Cb>神经场景表示与渲染（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fdeepmind.com\u002Fblog\u002Fneural-scene-representation-and-rendering\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_744614b07eb7.png\" \u002F>\u003C\u002Fp>\n\n:pill: \u003Cb>Im2Struct：从单张RGB图像恢复3D形状结构（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.05469.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e220612e0701.jpg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>FoldingNet：基于深度网格变形的点云自编码器（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.07262.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1c68aa17af73.jpg\" \u002F>\u003C\u002Fp>\n\n:camera::space_invader: \u003Cb>Pix3D：单张图像3D形状建模的数据集与方法（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fpix3d.csail.mit.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9594d245952d.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>3D-RCNN：基于渲染与比较的实例级3D物体重建（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002FCameraReady\u002F1128.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3911fb42933a.jpeg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>套娃网络：通过嵌套形状层预测3D几何形状（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.10975.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_48312309cc04.jpeg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>基于图卷积自编码器的可变形形状补全（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.00268v1.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_89cec1fa7d46.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>3D形状的全局到局部生成模型（SIGGRAPH Asia 2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fvcc.szu.edu.cn\u002Fresearch\u002F2018\u002FG2L)[[代码]](https:\u002F\u002Fgithub.com\u002FHao-HUST\u002FG2LGAN)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_53cd78c9e178.jpg\" \u002F>\u003C\u002Fp>\n\n:gem::game_die::space_invader: \u003Cb>ALIGNet：通过无监督学习实现部分形状无关对齐（TOG 2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fbit.ly\u002Falignet) [[代码]](https:\u002F\u002Fgithub.com\u002Franahanocka\u002FALIGNet\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5e69e9894f0b.png\" \u002F>\u003C\u002Fp>\n\n:game_die::space_invader: \u003Cb>PointGrid：用于3D形状理解的深度网络（CVPR 2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FLe_PointGrid_A_Deep_CVPR_2018_paper.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Ftrucleduc\u002FPointGrid)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5ae452ff2295.jpeg\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>GAL：用于单视角3D物体重建的几何对抗损失（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fxjqi.github.io\u002FGAL.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_aa0abf6416cc.gif\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>视觉对象网络：具有解耦3D表示的图像生成（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F7297-visual-object-networks-image-generation-with-disentangled-3d-representations.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_04cdceb5b42a.jpeg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>学习推断并执行3D形状程序（2019）\u003C\u002Fb> [[论文]](http:\u002F\u002Fshape2prog.csail.mit.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_068450740104.jpg\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>学习推断并执行3D形状程序（2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1901.05103.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f1c282bf915c.jpg\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>为单视角3D重建学习视图先验（CVPR 2019）\u003C\u002Fb> [[论文]](http:\u002F\u002Fhiroharu-kato.com\u002Fprojects_en\u002Fview_prior_learning.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_3317f3e24c94.png\" \u002F>\u003C\u002Fp>\n\n:gem::game_die: \u003Cb>使用二次损失学习3D模型的嵌入（BMVC 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1907.10250) [[代码]](https:\u002F\u002Fgithub.com\u002Fnitinagarwal\u002FQuadricLoss)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.ics.uci.edu\u002F~agarwal\u002Fbmvc_2019.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>CompoNet：通过部件合成与组合学习生成未见对象（ICCV 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.07441)[[代码]](https:\u002F\u002Fgithub.com\u002Fnschor\u002FCompoNet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_474e21cb705e.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>CoMA：卷积网格自编码器（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F439\u002F1285.pdf)[[代码（TF）]](https:\u002F\u002Fgithub.com\u002Fanuragranj\u002Fcoma)[[代码（PyTorch）]](https:\u002F\u002Fgithub.com\u002Fpixelite1201\u002Fpytorch_coma\u002F)[[代码（PyTorch）]](https:\u002F\u002Fgithub.com\u002Fsw-gong\u002Fcoma)\n\u003Cbr>[CoMA](https:\u002F\u002Fcoma.is.tue.mpg.de\u002F) 是一种多功能模型，它利用网格表面上的谱卷积学习人脸的非线性表示。CoMA 引入了网格采样操作，从而实现分层的网格表示，能够在模型内部以多尺度捕捉形状和表情中的非线性变化。\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Fcoma.is.tue.mpg.de\u002Fuploads\u002Fckeditor\u002Fpictures\u002F91\u002Fcontent_coma_faces.jpg\">\u003C\u002Fp>\n\n\u003Cb>RingNet：单张图像的三维人脸重建（2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F509\u002Fpaper_camera_ready.pdf)[[代码]](https:\u002F\u002Fgithub.com\u002Fsoubhiksanyal\u002FRingNet)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_179e4f2a621f.gif\">\u003C\u002Fp>\n\n\u003Cb>VOCA：语音驱动的角色动画（2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Fps.is.tuebingen.mpg.de\u002Fuploads_file\u002Fattachment\u002Fattachment\u002F510\u002Fpaper_final.pdf)[[视频]](https:\u002F\u002Fyoutu.be\u002FXceCxf_GyW4)[[代码]](https:\u002F\u002Fgithub.com\u002FTimoBolkart\u002Fvoca)\n\u003Cbr>[VOCA](https:\u002F\u002Fvoca.is.tue.mpg.de\u002F) 是一个简单且通用的语音驱动面部动画框架，适用于多种不同身份的人脸。该代码库展示了如何根据任意语音信号和静态角色网格来合成逼真的角色动画。\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1cdee30e402b.gif\">\u003C\u002Fp>\n\n:gem: \u003Cb>基于插值的可微渲染器学习预测三维物体\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.01210)[[网站]](https:\u002F\u002Fnv-tlabs.github.io\u002FDIB-R\u002F)[[代码]](https:\u002F\u002Fgithub.com\u002Fnv-tlabs\u002FDIB-R)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_56662af983c6.png\"> \u003C\u002Fp>\n\n:gem: \u003Cb>软光栅化器：用于基于图像的三维推理的可微渲染器\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.01786)[[代码]](https:\u002F\u002Fgithub.com\u002FShichenLiu\u002FSoftRas)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2b8d73b7f7ee.png\"> \u003C\u002Fp>\n\n\u003Cb>NeRF：将场景表示为神经辐射场以进行视图合成\u003C\u002Fb> [[项目]](http:\u002F\u002Fwww.matthewtancik.com\u002Fnerf)[[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2003.08934)[[代码]](https:\u002F\u002Fgithub.com\u002Fbmild\u002Fnerf)\n\u003Cp align=\"center\"> \u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ac3328a410fc.png\"> \u003C\u002Fp>\n\n:gem::game_die: \u003Cb>GAMesh：面向深度点网络的引导与增强网格化方法（3DV 2020）\u003C\u002Fb> [[项目]](https:\u002F\u002Fwww.ics.uci.edu\u002F~agarwal\u002FGAMesh\u002F)[[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2010.09774)[[代码]](https:\u002F\u002Fgithub.com\u002Fnitinagarwal\u002FGAMesh)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.ics.uci.edu\u002F~agarwal\u002F3DV_2020.png\" \u002F>\u003C\u002Fp>\n\n\n\n:space_invader: \u003Cb>生成式体素网：学习基于能量的模型用于三维形状的合成与分析（2020 TPAMI）\u003C\u002Fb>   [[论文]](http:\u002F\u002Fwww.stat.ucla.edu\u002F~jxie\u002F3DEBM\u002F3DEBM_file\u002Fdoc\u002FgVoxelNet.pdf) \n\n本文提出了一种深度三维能量模型来表示体素化的形状。该模型的最大似然训练遵循“由合成推导”的方案。实验表明，所提出的模型能够生成高质量的三维形状模式，并可用于各种三维形状分析任务。\n\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ed1b54811bc5.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>生成式点云网：面向无序点集的深度能量学习，用于三维生成、重建和分类（2021 CVPR）\u003C\u002Fb> [[项目]](http:\u002F\u002Fwww.stat.ucla.edu\u002F~jxie\u002FGPointNet\u002F)[[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2004.01301.pdf)[[代码]](https:\u002F\u002Fgithub.com\u002Ffei960922\u002FGPointNet)\n\n生成式点云网是一种基于能量的无序点云模型，其能量函数由一个输入排列不变的自底向上神经网络参数化。该模型可以通过基于 MCMC 的最大似然学习进行训练，也可以通过短时间运行的 MCMC 将能量模型作为类似流的生成器用于点云的重建和插值。学习到的点云表示对于点云分类非常有用。\n\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_249e4057d91d.png\" \u002F>\u003C\u002Fp>\n\n:game_die: :gem: \u003Cb>塑造我的脸：基于面到面变换的三维人脸扫描配准\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2012.09235)[[代码]](https:\u002F\u002Fgithub.com\u002Fmbahri\u002Fsmf)\n\n“塑造我的脸”（SMF）是一个点云到网格的自编码器，用于原始人脸扫描的配准以及合成人脸的生成。SMF 利用经过改进的 PointNet 编码器，结合视觉注意力模块和可微表面采样，使其不依赖于原始的表面表示，从而减少预处理的需求。网格卷积解码器与专门针对口腔的 PCA 模型相结合，并根据测地距离平滑融合，形成一个紧凑且对噪声具有高度鲁棒性的模型。SMF 被应用于在野外使用 iPhone 深度相机捕获的扫描数据的配准及表情迁移，这些数据既可以表示为网格，也可以表示为点云。\n\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_901644743a2c.png\" \u002F>\u003C\u002Fp>\n\n:game_die: \u003Cb>学习隐式场用于生成式形状建模（2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.02822)[[代码]](https:\u002F\u002Fgithub.com\u002Ftimzhang642\u002F3D-Machine-Learning)\n\n我们提倡使用隐式场来学习形状的生成模型，并引入了一种名为 IM-NET 的隐式场解码器，用于形状生成，旨在提升生成形状的视觉质量。隐式场为三维空间中的每个点分配一个值，从而可以将形状提取为等值面。IM-NET 经过训练，能够通过二元分类器完成这一分配。具体而言，它接收一个点的坐标以及编码形状特征的向量，然后输出一个值，指示该点是在形状外部还是内部。通过用我们的隐式解码器替代传统的解码器来进行表征学习（通过 IM-AE）和形状生成（通过 IM-GAN），我们在生成式形状建模、插值以及单目三维重建等任务中都取得了更优的结果，尤其是在视觉质量方面。\n\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4d6adb9307d1.png\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"material_synthesis\" \u002F>\n\n\n\n## 纹理\u002F材质分析与合成\n\u003Cb>使用卷积神经网络的纹理合成（2015）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1505.07376.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_af4589e38df6.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>用于静止材质的两步SVBRDF捕获（SIGGRAPH 2015）\u003C\u002Fb> [[论文]](https:\u002F\u002Fmediatech.aalto.fi\u002Fpublications\u002Fgraphics\u002FTwoShotSVBRDF\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ac53134c8c13.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于神经纹理合成的反射率建模（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Fmediatech.aalto.fi\u002Fpublications\u002Fgraphics\u002FNeuralSVBRDF\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7d543f74a91b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>利用自增强卷积神经网络从单张照片建模表面外观（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fmsraig.info\u002F~sanet\u002Fsanet.htm)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fmsraig.info\u002F~sanet\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>高分辨率多尺度神经纹理合成（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwxs.ca\u002Fresearch\u002Fmultiscale-neural-synthesis\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_bc1f283c0706.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>利用深度学习从单一材质的镜面物体中恢复反射率和自然光照（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Fhomes.cs.washington.edu\u002F~krematas\u002FPublications\u002Freflectance-natural-illumination.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.vision.ee.ethz.ch\u002F~georgous\u002Fimages\u002Ftpami17_teaser2.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>从野外照片集中联合估计材质与光照（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.08313.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2063f7bb2d56.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>J相机周围有什么？（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1611.09325v2.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fhomes.cs.washington.edu\u002F~krematas\u002Fmy_images\u002Farxiv16b_teaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>TextureGAN：用纹理块控制深度图像合成（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1706.02823.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Ftexturegan.eye.gatech.edu\u002Fimg\u002Fpaper_figure.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>高斯材质合成（2018 SIGGRAPH）\u003C\u002Fb> [[论文]](https:\u002F\u002Fusers.cg.tuwien.ac.at\u002Fzsolnai\u002Fgfx\u002Fgaussian-material-synthesis\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f4018a23c97e.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>通过对抗扩张实现非平稳纹理合成（2018 SIGGRAPH）\u003C\u002Fb> [[论文]](http:\u002F\u002Fvcc.szu.edu.cn\u002Fresearch\u002F2018\u002FTexSyn)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_20e5cdcebc0f.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>通过图像及梯度幅值系数的多尺度空间与统计纹理特征评估合成纹理质量（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.08020.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0f2e127a0441.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>LIME：实时内在材质估计（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Fgvv.mpi-inf.mpg.de\u002Fprojects\u002FLIME\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fweb.stanford.edu\u002F~zollhoef\u002Fpapers\u002FCVPR18_Material\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>利用渲染感知深度网络进行单张图像SVBRDF捕获（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fteam.inria.fr\u002Fgraphdeco\u002Ffr\u002Fprojects\u002Fdeep-materials\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7e4b3b0a498a.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PhotoShape：用于大规模形状集合的逼真材质（2018）\u003C\u002Fb> [[论文]](https:\u002F\u002Fkeunhong.com\u002Fpublications\u002Fphotoshape\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fkeunhong.com\u002Fpublications\u002Fphotoshape\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>为3D形状学习材质感知局部描述符（2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.vovakim.com\u002Fpapers\u002F18_3DV_ShapeMatFeat.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_731bd52f99dd.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>FrankenGAN：使用风格同步GAN引导建筑体量模型的细节合成（2018 SIGGRAPH Asia）\u003C\u002Fb> [[论文]](http:\u002F\u002Fgeometry.cs.ucl.ac.uk\u002Fprojects\u002F2018\u002Ffrankengan\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a4820745e08e.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"style_transfer\" \u002F>\n\n## 风格学习与迁移\n\u003Cb>基于各向异性部件尺度的风格-内容分离（2010）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.cs.sfu.ca\u002F~haoz\u002Fpubs\u002Fxu_siga10_style.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ed473ab4310b.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>保持设计的服装迁移（2012）\u003C\u002Fb> [[论文]](https:\u002F\u002Fhal.inria.fr\u002Fhal-00695903\u002Ffile\u002FGarmentTransfer.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2f8a45988aba.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>类比驱动的三维风格迁移（2014）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.chongyangma.com\u002Fpublications\u002Fst\u002Findex.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_adc2b4646158.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>风格要素：学习感知形状风格相似性（2015）\u003C\u002Fb> [[论文]](http:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FStyleSimilarity\u002FStyleSimilarity.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fhappylun\u002FStyleSimilarity)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FStyleSimilarity\u002FStyleSimilarity_teaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>保持功能性的形状风格迁移（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FStyleTransfer\u002FStyleTransfer.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fhappylun\u002FStyleTransfer)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fpeople.cs.umass.edu\u002F~zlun\u002Fpapers\u002FStyleTransfer\u002FStyleTransfer_teaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>无监督的图像到模型集合的纹理迁移（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fai.stanford.edu\u002F~haosu\u002Fpapers\u002Fsiga16_texture_transfer_small.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9cec00d0dd45.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于几何特征的学习细节迁移（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fsurfacedetails.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f92955f96de0.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>三维形状上风格定义元素的协同定位（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fpeople.scs.carleton.ca\u002F~olivervankaick\u002Fpubs\u002Fstyle_elem.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_1efe9fed95b9.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>神经网络三维网格渲染器（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fhiroharu-kato.com\u002Fprojects_en\u002Fneural_renderer.html) [[代码]](https:\u002F\u002Fgithub.com\u002Fhiroharu-kato\u002Fneural_renderer.git)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7432f17d2162.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>通过代理与图像对齐进行外观建模（2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fvcc.szu.edu.cn\u002Fresearch\u002F2018\u002FAppMod)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_65b1208a46b1.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>Pixel2Mesh：从单张RGB图像生成三维网格模型（2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fbigvid.fudan.edu.cn\u002Fpixel2mesh\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5ee4816688b0.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>自动非配对形状变形迁移（SIGGRAPH Asia 2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fgeometrylearning.com\u002Fausdt\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9ac6e7229911.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3DSNet：无监督的形状到形状三维风格迁移（2020）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2011.13388) [[代码]](https:\u002F\u002Fgithub.com\u002Fethz-asl\u002F3dsnet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_b64e08b146a8.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Ca name=\"scene_synthesis\" \u002F>\n\n## 场景合成\u002F重建\n\u003Cb>让家更温馨：家具布局的自动优化（2011，SIGGRAPH）\u003C\u002Fb> [[论文]](http:\u002F\u002Fpeople.sutd.edu.sg\u002F~saikit\u002Fprojects\u002Ffurniture\u002Findex.html)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7a7e45d17486.gif\" \u002F>\u003C\u002Fp>\n\n\u003Cb>利用室内设计指南进行交互式家具布局（2011）\u003C\u002Fb> [[论文]](http:\u002F\u002Fgraphics.stanford.edu\u002F~pmerrell\u002FfurnitureLayout.htm)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_072812b3709f.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>使用局部退火可逆跳跃MCMC在约束条件下合成开放世界（2012）\u003C\u002Fb> [[论文]](http:\u002F\u002Fgraphics.stanford.edu\u002F~lfyg\u002Fowl.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f698fc8e9758.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于示例的三维物体布局合成（2012 SIGGRAPH Asia）\u003C\u002Fb> [[论文]](http:\u002F\u002Fgraphics.stanford.edu\u002Fprojects\u002Fscenesynth\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dd61aa2cc1f0.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Skyetch2Scene：基于草图的三维模型协同检索与协同放置（2013）\u003C\u002Fb> [[论文]](http:\u002F\u002Fsweb.cityu.edu.hk\u002Fhongbofu\u002Fprojects\u002Fsketch2scene_sig13\u002F#.WWWge__ysb0)\n\u003Cp align=\"center\">\u003Cimg width=\"40%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_801103499af0.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>动作驱动的三维室内场景演化（2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.cs.sfu.ca\u002F~haoz\u002Fpubs\u002Fma_siga16_action.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2b82305e6be2.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>The Clutterpalette：用于细化室内场景的交互式工具（2015）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.cs.umb.edu\u002F~craigyu\u002Fpapers\u002Fclutterpalette.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7352fdc32356.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Image2Scene：改变三维房间的风格（2015）\u003C\u002Fb> [[论文]](https:\u002F\u002Fdl.acm.org\u002Fdoi\u002Fabs\u002F10.1145\u002F2733373.2806274)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e1de6c4e58a1.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>用于创建场景变体的关系模板（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fgeometry.cs.ucl.ac.uk\u002Fprojects\u002F2016\u002Frelationship-templates\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_fdeaa4ef954b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>IM2CAD（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fhomes.cs.washington.edu\u002F~izadinia\u002Fim2cad.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fi.imgur.com\u002FKhtOeuB.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>预测完整的室内场景三维模型（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1504.02437.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_0c429d0e0152.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>从单张RGBD图像完整解析三维场景（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.09490.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_8d4ca3868c4f.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>栅格转矢量：重新审视平面图转换（2017年，ICCV）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.cse.wustl.edu\u002F~chenliu\u002Ffloorplan-transformation.html) [[代码]](https:\u002F\u002Fgithub.com\u002Fart-programmer\u002FFloorplanTransformation)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.cse.wustl.edu\u002F~chenliu\u002Ffloorplan-transformation\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>用于3D多物体场景的全卷积精炼自编码生成对抗网络（2017年）\u003C\u002Fb> [[博客]](https:\u002F\u002Fbecominghuman.ai\u002F3d-multi-object-gan-7b7cee4abf80)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_eeba5f7ae75f.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于活动关联对象关系图的室内场景自适应合成（2017年SIGGRAPH Asia）\u003C\u002Fb> [[论文]](http:\u002F\u002Farts.buaa.edu.cn\u002Fprojects\u002Fsa17\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_23f0b545b463.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>利用遗传算法的自动化室内设计（2017年）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpublik.tuwien.ac.at\u002Ffiles\u002Fpublik_262718.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.peterkan.com\u002Fpictures\u002Fteaserq.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SceneSuggest：情境驱动的3D场景设计（2017年）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1703.00061.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_211125dbed0b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>一种端到端深度学习方法，用于实时同步的3D重建与材料识别（2017年）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1703.04699v1.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_7ac82b0fbcfb.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于随机文法的人本化室内场景合成（2018年，CVPR）\u003C\u002Fb>[[论文]](http:\u002F\u002Fweb.cs.ucla.edu\u002F~syqi\u002Fpublications\u002Fcvpr2018synthesis\u002Fcvpr2018synthesis.pdf) [[补充材料]](http:\u002F\u002Fweb.cs.ucla.edu\u002F~syqi\u002Fpublications\u002Fcvpr2018synthesis\u002Fcvpr2018synthesis_supplementary.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002FSiyuanQi\u002Fhuman-centric-scene-synthesis)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fweb.cs.ucla.edu\u002F~syqi\u002Fpublications\u002Fthumbnails\u002Fcvpr2018synthesis.gif\" \u002F>\u003C\u002Fp>\n\n:camera::game_die: \u003Cb>FloorNet：基于3D扫描的平面图重建统一框架（2018年）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.00090.pdf) [[代码]](http:\u002F\u002Fart-programmer.github.io\u002Ffloornet.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f8e0766d7c5a.png\" \u002F>\u003C\u002Fp>\n\n:space_invader: \u003Cb>ScanComplete：大规模场景补全与3D扫描的语义分割（2018年）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.10215.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_404cfe20d42e.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>用于室内场景合成的深度卷积先验（2018年）\u003C\u002Fb> [[论文]](https:\u002F\u002Fkwang-ether.github.io\u002Fpdf\u002Fdeepsynth.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5a365413a3a4.png\" \u002F>\u003C\u002Fp>\n\n:camera: \u003Cb>基于深度卷积生成模型的快速灵活室内场景合成（2018年）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1811.12463.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fbrownvc\u002Ffast-synth)\n\u003Cp align=\"center\">\u003Cimg width=\"80%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ed86d51824eb.jpg\" >\u003C\u002Fp>\n\n\u003Cb>使用随机文法实现可配置的3D场景合成与逐像素真值的2D图像渲染（2018年）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1704.00112.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_19fe0afbb156.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>从单张RGB图像进行整体3D场景解析与重建（ECCV 2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fsiyuanhuang.com\u002Fholistic_parsing\u002Fmain.html) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fweb.cs.ucla.edu\u002F~syqi\u002Fpublications\u002Fthumbnails\u002Feccv2018scene.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于语言的场景数据库驱动的3D场景合成（SIGGRAPH Asia 2018）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.sfu.ca\u002F~agadipat\u002Fpublications\u002F2018\u002FT2S\u002Fproject_page.html) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.sfu.ca\u002F~agadipat\u002Fpublications\u002F2018\u002FT2S\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于混合表示的场景合成深度生成建模（2018年）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1808.02084.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_14e57b867c41.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>GRAINS：面向室内场景的生成式递归自编码器（2018年）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1807.09193.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.groundai.com\u002Fmedia\u002Farxiv_projects\u002F373503\u002Fnew_pics\u002Fteaserfig.jpg.750x0_q75_crop.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SEETHROUGH：在严重遮挡的室内场景图像中寻找物体（2018年）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.vovakim.com\u002Fpapers\u002F18_3DVOral_SeeThrough.pdf) \n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4f533697c97d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:space_invader: Scan2CAD：在RGB-D扫描中学习CAD模型对齐（CVPR 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1811.11187.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fskanti\u002FScan2CAD)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d0c60afde578.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:gem: Scan2Mesh：从非结构化距离扫描到3D网格（CVPR 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1811.10464.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f5ac066de6e4.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:space_invader: 3D-SIC：针对RGB-D扫描的3D语义实例补全（arXiv 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1904.12012.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.niessnerlab.org\u002Fpapers\u002F2019\u002Fz1sic\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:space_invader: 3D扫描中的端到端CAD模型检索与9自由度对齐（arXiv 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1906.04201)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fwww.niessnerlab.org\u002Fpapers\u002F2019\u002Fz2end2end\u002Fteaser.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>3D室内场景合成综述（2020年）\u003C\u002Fb> [[论文]](https:\u002F\u002Fwww.researchgate.net\u002Fprofile\u002FShao_Kui_Zhang\u002Fpublication\u002F333135099_A_Survey_of_3D_Indoor_Scene_Synthesis\u002Flinks\u002F5ce13a5492851c4eabad4de0\u002FA-Survey-of-3D-Indoor-Scene-Synthesis.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f80396501ca9.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:pill: :camera: PlanIT：利用关系图和空间先验网络进行室内场景规划与实例化（2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Fkwang-ether.github.io\u002Fpdf\u002Fplanit.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fbrownvc\u002Fplanit)\n\u003Cp align=\"center\">\u003Cimg src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_63f511aed36a.jpg\">\u003C\u002Fp>\n\n\u003Cb>:space_invader: 特征度量配准：一种无需对应点的快速半监督鲁棒点云配准方法（CVPR 2020）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F2005.01014)[[代码]](https:\u002F\u002Fgithub.com\u002FXiaoshuiHuang\u002Ffmr)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fgithub.com\u002FXiaoshuiHuang\u002Fxiaoshuihuang.github.io\u002Fblob\u002Fmaster\u002Fresearch\u002F2020-feature-metric.png?raw=true\" \u002F>\u003C\u002Fp>\n\n\u003Cb>:pill: 面向人的室内场景评估与合成指标（2020）\u003C\u002Fb> [[论文]](sciencedirect.com\u002Fscience\u002Farticle\u002Fabs\u002Fpii\u002FS1524070320300175)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_48a06950d110.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SceneCAD：预测RGB-D扫描中的物体对齐与布局（2020）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F2003.12622.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_dbdaef7a5b20.jpg\" \u002F>\u003C\u002Fp>\n\n\n\n\n\n\u003Ca name=\"scene_understanding\" \u002F>\n\n\n\n## 场景理解（另一个更详细的[仓库](https:\u002F\u002Fgithub.com\u002Fbertjiazheng\u002Fawesome-scene-understanding)）\n\n\u003Cb>恢复杂乱房间的空间布局（2009）\u003C\u002Fb> [[论文]](http:\u002F\u002Fdhoiem.cs.illinois.edu\u002Fpublications\u002Ficcv2009_hedau_indoor.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_cdacc16c94b7.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>利用图核刻画场景中的结构关系（2011 SIGGRAPH）\u003C\u002Fb> [[论文]](https:\u002F\u002Fgraphics.stanford.edu\u002F~mdfisher\u002FgraphKernel.html)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f8389736e53b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>使用3D几何短语理解室内场景（2013）\u003C\u002Fb> [[论文]](http:\u002F\u002Fcvgl.stanford.edu\u002Fprojects\u002F3dgp\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_a7e0b8ec2241.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>通过情境焦点组织异构场景集合（2014 SIGGRAPH）\u003C\u002Fb> [[论文]](http:\u002F\u002Fkevinkaixu.net\u002Fprojects\u002Ffocal.html)\n\u003Cp align=\"center\">\u003Cimg width=\"60%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_2977e6dc4db0.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SceneGrok：在3D环境中推断动作地图（2014，SIGGRAPH）\u003C\u002Fb> [[论文]](http:\u002F\u002Fgraphics.stanford.edu\u002Fprojects\u002Fscenegrok\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_771bfa73c82d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PanoContext：用于全景场景理解的全室3D上下文模型（2014）\u003C\u002Fb> [[论文]](http:\u002F\u002Fpanocontext.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_9e6007d00ea6.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>学习信息丰富的边缘图用于室内场景布局预测（2015）\u003C\u002Fb> [[论文]](http:\u002F\u002Fslazebni.cs.illinois.edu\u002Fpublications\u002Ficcv15_informative.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_defcfa4250a0.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Rent3D：用于单目布局估计的平面图先验（2015）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.cs.toronto.edu\u002F~fidler\u002Fprojects\u002Frent3D.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_efc4193b0dc7.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>一种由粗到细的室内布局估计方法（CFILE，2016）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpdfs.semanticscholar.org\u002F7024\u002Fa92186b81e6133dc779f497d06877b48d82b.pdf?_ga=2.54181869.497995160.1510977308-665742395.1510465328)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4d3964f8209d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>DeLay：针对杂乱室内场景的鲁棒空间布局估计（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fdeeplayout.stanford.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"30%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_14b0542c8f4d.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>大规模室内空间的3D语义解析（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fbuildingparser.stanford.edu\u002Fmethod.html) [[代码]](https:\u002F\u002Fgithub.com\u002Falexsax\u002F2D-3D-Semantics)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"http:\u002F\u002Fbuildingparser.stanford.edu\u002Fimages\u002Fteaser.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>单张图像3D解释网络（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002F3dinterpreter.csail.mit.edu\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fjiajunwu\u002F3dinn)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_26d9f64f28e6.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>深度多模态图像对应学习（2016）\u003C\u002Fb> [[论文]](http:\u002F\u002Fwww.cse.wustl.edu\u002F~chenliu\u002Ffloorplan-matching.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ee304c9d0d94.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于物理的渲染结合卷积神经网络用于室内场景理解（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002F3dvision.princeton.edu\u002Fprojects\u002F2016\u002FPBRS\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fyindaz\u002Fpbrs) [[代码]](https:\u002F\u002Fgithub.com\u002Fyindaz\u002Fsurface_normal) [[代码]](https:\u002F\u002Fgithub.com\u002Ffyu\u002Fdilation) [[代码]](https:\u002F\u002Fgithub.com\u002Fs9xie\u002Fhed)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_e72a7e8c1109.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>RoomNet：端到端房间布局估计（2017）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1703.06241.pdf)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_d40352372be3.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>SUN RGB-D：一个RGB-D场景理解基准套件（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Frgbd.cs.princeton.edu\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_205289780305.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>从单张深度图像进行语义场景补全（2017）\u003C\u002Fb> [[论文]](http:\u002F\u002Fsscnet.cs.princeton.edu\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fshurans\u002Fsscnet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_529aad9e6055.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>从3D场景的2D图像中分解形状、姿态和布局（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1712.01812.pdf) [[代码]](https:\u002F\u002Fshubhtuls.github.io\u002Ffactored3d\u002F)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_f5019df0ba20.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>LayoutNet：从单张RGB图像重建3D房间布局（2018 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1803.08999.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fzouchuhang\u002FLayoutNet)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_c07b707be767.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PlaneNet：基于单张RGB图像的分段平面重建（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fart-programmer.github.io\u002Fplanenet\u002Fpaper.pdf) [[代码]](http:\u002F\u002Fart-programmer.github.io\u002Fplanenet.html)\n\u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4869b4f17c23.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>利用合成图像进行跨域自监督多任务特征学习（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fweb.cs.ucdavis.edu\u002F~yjlee\u002Fprojects\u002Fcvpr2018.pdf) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fjason718.github.io\u002Fproject\u002Fcvpr18\u002Ffiles\u002Fconcept_pic.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Pano2CAD：从单张全景图像中恢复房间布局（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fbjornstenger.github.io\u002Fpapers\u002Fxu_wacv2017.pdf) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fwww.groundai.com\u002Fmedia\u002Farxiv_projects\u002F58924\u002Ffigures\u002FCompare_2b.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于单张全景图的室内场景自动三维建模（2018 CVPR）\u003C\u002Fb> [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FYang_Automatic_3D_Indoor_CVPR_2018_paper.pdf) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_aad812281fca.jpeg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于关联嵌入的单幅图像分段平面三维重建（2019 CVPR）\u003C\u002Fb> [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1902.09777.pdf) [[代码]](https:\u002F\u002Fgithub.com\u002Fsvip-lab\u002FPlanarReconstruction) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_5b7e9eb18828.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>基于逆向图形的三维感知场景操控（NeurIPS 2018）\u003C\u002Fb> [[论文]](http:\u002F\u002F3dsdn.csail.mit.edu\u002F) [[代码]](https:\u002F\u002Fgithub.com\u002Fsvip-lab\u002FPlanarReconstruction) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_bf98ad17af78.png\" \u002F>\u003C\u002Fp>\n\n:gem: \u003Cb>基于多层深度与极线变换器的三维场景重建（ICCV 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Fresearch.dshin.org\u002Ficcv19\u002Fmulti-layer-depth\u002F) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_ec94229c38af.png\" \u002F>\u003Cbr>\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_15e02a6157b3.jpg\" \u002F>\u003C\u002Fp>\n\n\u003Cb>PerspectiveNet：通过透视点实现单张RGB图像中的三维目标检测（NIPS 2019）\u003C\u002Fb> [[论文]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F9093-perspectivenet-3d-object-detection-from-a-single-rgb-image-via-perspective-points.pdf) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Fstorage.googleapis.com\u002Fgroundai-web-prod\u002Fmedia\u002Fusers\u002Fuser_288036\u002Fproject_402358\u002Fimages\u002Fx1.png\" \u002F>\u003C\u002Fp>\n\n\u003Cb>Holistic++ 场景理解：单视角三维整体场景解析与人体姿态估计，包含人-物交互及物理常识推理（ICCV 2019）\u003C\u002Fb> [[论文与代码]](https:\u002F\u002Fgithub.com\u002Fyixchen\u002Fholistic_scene_human) \u003Cp align=\"center\">\u003Cimg width=\"50%\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_readme_4795201634cb.png\" \u002F>\u003C\u002Fp>","# 3D-Machine-Learning 快速上手指南\n\n**注意**：`3D-Machine-Learning` 并非一个可直接安装运行的软件包或库，而是一个** curated（精选）的资源汇总仓库**。它主要收集了 3D 机器学习领域的课程、数据集、论文分类及研究笔记。因此，本指南将指导你如何获取该资源列表，并基于其中的内容搭建开发环境。\n\n## 1. 环境准备\n\n由于本项目是资源索引，你需要根据你选择的具体研究方向（如点云处理、3D 重建等）配置相应的深度学习环境。以下是通用的基础环境要求：\n\n*   **操作系统**: Linux (推荐 Ubuntu 18.04\u002F20.04) 或 macOS\n*   **编程语言**: Python 3.7+\n*   **核心依赖**:\n    *   PyTorch 或 TensorFlow (根据具体论文代码要求)\n    *   CUDA & cuDNN (用于 GPU 加速)\n*   **常用 3D 数据处理库** (建议预先安装):\n    ```bash\n    pip install numpy scipy matplotlib trimesh open3d pytorch3d\n    ```\n\n> **国内加速建议**：\n> 推荐使用清华或中科大镜像源安装 Python 依赖，以提升下载速度：\n> ```bash\n> pip install -i https:\u002F\u002Fpypi.tuna.tsinghua.edu.cn\u002Fsimple \u003Cpackage_name>\n> ```\n\n## 2. 获取资源与安装步骤\n\n你不需要“安装”此工具，而是需要克隆仓库以获取最新的数据集链接、课程笔记和论文分类。\n\n### 步骤 1: 克隆仓库\n使用 Git 将资源库下载到本地：\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Ftimzhang642\u002F3D-Machine-Learning.git\ncd 3D-Machine-Learning\n```\n\n> **国内加速建议**：\n> 如果 GitHub 连接缓慢，可使用 Gitee 镜像（如有）或通过代理加速：\n> ```bash\n> git clone https:\u002F\u002Fgitee.com\u002Fmirror\u002F3D-Machine-Learning.git\n> # 或者使用国内代理前缀\n> git clone https:\u002F\u002Fghproxy.com\u002Fhttps:\u002F\u002Fgithub.com\u002Ftimzhang642\u002F3D-Machine-Learning.git\n> ```\n\n### 步骤 2: 浏览资源目录\n进入目录后，直接查看 `README.md` 文件，其中包含了按领域分类的详细资源列表：\n*   **Courses**: 斯坦福、MIT 等名校的 3D 视觉课程。\n*   **Datasets**: ModelNet, ShapeNet, ScanObjectNN 等主流数据集下载链接。\n*   **Tasks**: 涵盖分类、检测、分割、重建、生成等任务的论文索引。\n\n```bash\ncat README.md\n```\n\n## 3. 基本使用\n\n本项目的核心用法是**作为研究导航**，引导你找到特定的数据集或论文代码。以下是典型的使用流程示例：\n\n### 场景：寻找点云分类数据集并加载示例\n\n假设你想进行点云分类研究，根据仓库中的 [Datasets](#datasets) 章节，你发现了 `ScanObjectNN` 数据集。\n\n1.  **访问链接**：在 README 中找到 `ScanObjectNN` 的链接 (https:\u002F\u002Fhkust-vgd.github.io\u002Fscanobjectnn\u002F) 并下载数据。\n2.  **使用 Open3D 加载数据** (示例代码)：\n    下载完成后，你可以使用 Python 和 `open3d` 库快速查看数据：\n\n    ```python\n    import open3d as o3d\n\n    # 替换为你实际下载的数据路径\n    file_path = \"path\u002Fto\u002Fscanobjectnn_data\u002Fexample.ply\"\n\n    # 读取点云\n    pcd = o3d.io.read_point_cloud(file_path)\n\n    # 可视化\n    o3d.visualization.draw_geometries([pcd])\n    ```\n\n### 场景：学习相关课程\n根据 [Courses](#courses) 章节，你可以直接访问斯坦福 CS231A 课程主页进行学习：\n*   访问：http:\u002F\u002Fweb.stanford.edu\u002Fclass\u002Fcs231a\u002F\n*   获取课件、作业和讲座视频以构建理论基础。\n\n### 参与社区\n如需进一步交流，可点击 README 中的 Slack 邀请链接加入全球 3D 机器学习社区，提问或分享研究成果。\n\n---\n*提示：具体的算法实现代码通常位于各篇论文对应的独立 GitHub 仓库中，请通过本仓库提供的论文链接跳转获取。*","一家自动驾驶初创公司的算法团队正急需构建一套能精准识别复杂路况中行人和车辆的 3D 感知系统，以优化其自动紧急制动功能。\n\n### 没有 3D-Machine-Learning 时\n- **资源搜集如大海捞针**：工程师需手动在 arXiv、Google Scholar 等多个平台搜索“点云检测”或\"3D 重建”论文，耗时数周仍难以覆盖最新成果。\n- **数据格式混乱难统一**：面对激光雷达生成的点云（:game_die:）与相机多视图图像（:camera:），团队缺乏明确的数据集指引，常因数据预处理标准不一导致模型训练失败。\n- **技术路线选择盲目**：由于缺乏对体素化（:space_invader:）与网格（:gem:）等不同 3D 表示方法的系统对比，团队在选型时反复试错，严重拖慢研发进度。\n- **学术脉络断裂**：难以理清经典算法与前沿研究之间的演进关系，导致复现代码时频繁遇到未记录的依赖问题或理论盲区。\n\n### 使用 3D-Machine-Learning 后\n- **一站式资源导航**：团队直接利用该仓库的分类目录，快速定位到\"Multiple Objects Detection\"板块，瞬间获取该领域最权威的论文列表与开源代码链接。\n- **精准匹配数据源**：通过\"Datasets\"章节，迅速找到适配点云与多视图融合任务的公开数据集（如 Princeton Shape Benchmark），大幅缩短数据清洗周期。\n- **清晰的技术图谱**：借助图标分类体系，工程师直观对比不同 3D 表示方法的优劣，迅速确定采用“基于深度学习的点云处理”作为核心路线。\n- **高效社区协作**：加入配套的 Slack 社区后，团队成员直接向全球专家请教几何深度学习难点，将原本需要数月摸索的调参过程压缩至几天。\n\n3D-Machine-Learning 将原本分散孤立的学术资源转化为结构化的研发加速器，帮助团队在竞争激烈的自动驾驶赛道上实现了从“盲目探索”到“精准打击”的跨越。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002Ftimzhang642_3D-Machine-Learning_dec5bbf4.jpg","timzhang642","Yuxuan (Tim) Zhang","https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002Ftimzhang642_a8c0194a.jpg","Visual storytelling for everyone.\r\n\r\nWe are hiring!",null,"timzhangyuxuan@gmail.com","ExperiencerTim","artflow.ai","https:\u002F\u002Fgithub.com\u002Ftimzhang642",10150,1808,"2026-04-03T18:31:09",5,"","未说明",{"notes":91,"python":89,"dependencies":92},"该仓库是一个用于整理和分类 3D 机器学习研究论文的资源列表（Awesome List），并非一个可执行的软件工具或代码库。因此，README 中未包含任何关于操作系统、GPU、内存、Python 版本或依赖库的安装运行需求。用户主要使用该页面查找数据集链接、课程资源和相关学术论文。",[],[54,13],[95,96,97,98,99,100,101,102,103,104],"3d-reconstruction","papers","neural-networks","3d","machine-learning","mesh","voxel","point-cloud","primitives","constructive-solid-geometries","2026-03-27T02:49:30.150509","2026-04-06T06:51:54.793438",[],[]]