[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"similar-becauseofAI--awesome-face":3,"tool-becauseofAI--awesome-face":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":76,"owner_location":76,"owner_email":76,"owner_twitter":76,"owner_website":76,"owner_url":79,"languages":80,"stars":93,"forks":94,"last_commit_at":95,"license":96,"difficulty_score":97,"env_os":98,"env_gpu":99,"env_ram":99,"env_deps":100,"category_tags":103,"github_topics":104,"view_count":23,"oss_zip_url":76,"oss_zip_packed_at":76,"status":16,"created_at":124,"updated_at":125,"faqs":126,"releases":127},3970,"becauseofAI\u002Fawesome-face","awesome-face","An awesome face technology repository.","awesome-face 是一个专注于人脸技术领域的开源资源合集，旨在为开发者和研究人员提供一站式的学术与工程参考。它系统性地整理了人脸识别、检测、关键点定位、聚类、表情分析、3D 重建以及生成对抗网络（GAN）应用（如换脸、老化模拟、去遮挡）等全方位的技术资料。\n\n面对人脸技术领域论文繁多、代码分散、难以追踪最新进展的痛点，awesome-face 通过持续更新，解决了信息碎片化的问题。它不仅收录了各类基准测试数据集和评估指标，还特别标注了在 WIDER Face、WFLW 和 MegaFace 等权威榜单上达到最先进水平（SOTA）的算法模型，帮助用户快速锁定高质量技术方案。\n\n该项目非常适合人工智能领域的科研人员、算法工程师以及计算机视觉开发者使用。无论是希望复现前沿论文的研究者，还是正在寻找成熟模块进行项目开发的工程师，都能从中高效获取所需资源。其独特的亮点在于对技术方向的细致分类以及对顶会成果（如 CVPR 论文）的及时跟进，让复杂的人脸技术生态变得清晰有序，是探索人脸科技不可或缺的导航工具。","# HelloFace [![Mentioned in Awesome HelloFace](https:\u002F\u002Fawesome.re\u002Fmentioned-badge.svg)](https:\u002F\u002Fgithub.com\u002FbecauseofAI\u002FHelloFace)   \nAn Awesome Face Technology Repository. (**Updating**)  \n\n## :trophy: SOTA\n[![PWC](https:\u002F\u002Fimg.shields.io\u002Fendpoint.svg?url=https:\u002F\u002Fpaperswithcode.com\u002Fbadge\u002Ftinaface-strong-but-simple-baseline-for-face\u002Fface-detection-on-wider-face-hard)](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fface-detection-on-wider-face-hard?p=tinaface-strong-but-simple-baseline-for-face)  \n[![PWC](https:\u002F\u002Fimg.shields.io\u002Fendpoint.svg?url=https:\u002F\u002Fpaperswithcode.com\u002Fbadge\u002Fsubpixel-heatmap-regression-for-facial\u002Fface-alignment-on-wflw)](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fface-alignment-on-wflw?p=subpixel-heatmap-regression-for-facial)  \n[![PWC](https:\u002F\u002Fimg.shields.io\u002Fendpoint.svg?url=https:\u002F\u002Fpaperswithcode.com\u002Fbadge\u002Fdeep-polynomial-neural-networks\u002Fface-verification-on-megaface)](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fface-verification-on-megaface?p=deep-polynomial-neural-networks)\n\n## :dart: Highlight\n\u003Cdiv align=\"center\">\u003Cimg width=\"1280\" height=\"auto\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FbecauseofAI_awesome-face_readme_d0ab636e646b.jpg\"\u002F>\u003C\u002Fdiv>\n\u003Cp align=\"center\">face detection\u003C\u002Fp>  \n\n\u003Cdiv align=\"center\">\u003Cimg width=\"1280\" height=\"auto\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FbecauseofAI_awesome-face_readme_fa47efb31ecd.jpg\"\u002F>\u003C\u002Fdiv>\n\u003Cp align=\"center\">face alignment\u003C\u002Fp>  \n\n\u003Cdiv align=\"center\">\u003Cimg width=\"1280\" height=\"auto\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FbecauseofAI_awesome-face_readme_37c1e099f77f.jpg\"\u002F>\u003C\u002Fdiv>\n\u003Cp align=\"center\">face recognition\u003C\u002Fp>  \n\n## :computer: Website\nhttps:\u002F\u002FbecauseofAI.github.io\u002FHelloFace (Welcome to maintain the website as a contributor through pulling request.)\n\n## :bookmark_tabs: Content\n- [Recent Update](#recent-update)\n  - [2022-04-05](#2022-04-05)\n  - [2020-10-02](#2020-10-02)\n  - [2020-01-26](#2020-01-26)\n  - [2019-07-11](#2019-07-11)\n  - [2019-04-06](#2019-04-06)\n  - [2019-01-12](#2019-01-12)\n  - [2018-12-01](#2018-12-01)\n  - [2018-07-21](#2018-07-21)\n  - [2018-04-20](#2018-04-20)\n  - [2018-03-28](#2018-03-28)\n- [Face Benchmark and Dataset](#face-benchmark-and-dataset)\n  - [Face Recognition Data](#face-recognition-data)\n  - [Face Detection Data](#face-detection-data)\n  - [Face Landmark Data](#face-landmark-data)\n  - [Face Attribute Data](#face-attribute-data)\n- [Face Recognition](#face-recognition)\n- [Face Detection](#face-detection)\n- [Face Landmark](#face-landmark)\n- [Face Clustering](#face-clustering)\n- [Face Expression](#face-expression)\n- [Face Action](#face-action)\n- [Face 3D](#face-3d)\n- [Face GAN](#face-gan)\n  - [Face Character](#face-character)\n  - [Face Editing](#face-editing)\n  - [Face De-Occlusion](#face-de-occlusion)\n  - [Face Aging](#face-aging)\n  - [Face Drawing](#face-drawing)\n  - [Face Generation](#face-generation)\n  - [Face Makeup](#face-makeup)\n  - [Face Swap](#face-swap)\n  - [Face Other](#face-other)\n- [Face Deblurring](#face-deblurring)\n- [Face Super-Resolution](#face-super-resolution)\n- [Face Manipulation](#face-manipulation)\n- [Face Anti-Spoofing](#face-anti-spoofing)\n- [Face Adversarial Attack](#face-adversarial-attack)\n- [Face Cross-Modal](#face-cross-modal)\n- [Face Capture](#face-capture)\n- [Face Lib and Tool](#face-lib-and-tool)\n\n## 👋 Recent Update\n###### 2022-04-05\n**CVPR2021**  \n- **VirFace**: Enhancing Face Recognition via Unlabeled Shallow Data [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_VirFace_Enhancing_Face_Recognition_via_Unlabeled_Shallow_Data_CVPR_2021_paper.pdf)  \n- **MagFace**: A Universal Representation for Face Recognition and Quality Assessment [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FMeng_MagFace_A_Universal_Representation_for_Face_Recognition_and_Quality_Assessment_CVPR_2021_paper.pdf)\n- Variational Prototype Learning for Deep Face Recognition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FDeng_Variational_Prototype_Learning_for_Deep_Face_Recognition_CVPR_2021_paper.pdf)\n- Cross-Domain Similarity Learning for Face Recognition in Unseen Domains [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FFaraki_Cross-Domain_Similarity_Learning_for_Face_Recognition_in_Unseen_Domains_CVPR_2021_paper.pdf)\n- Virtual Fully-Connected Layer: Training a Large-Scale Face Recognition Dataset With Limited Computational Resources [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Virtual_Fully-Connected_Layer_Training_a_Large-Scale_Face_Recognition_Dataset_With_CVPR_2021_paper.pdf)\n- Mitigating Face Recognition Bias via Group Adaptive Classifier [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FGong_Mitigating_Face_Recognition_Bias_via_Group_Adaptive_Classifier_CVPR_2021_paper.pdf)\n- Pseudo Facial Generation With Extreme Poses for Face Recognition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FWang_Pseudo_Facial_Generation_With_Extreme_Poses_for_Face_Recognition_CVPR_2021_paper.pdf)  \n- Dynamic Class Queue for Large Scale Face Recognition in the Wild [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Dynamic_Class_Queue_for_Large_Scale_Face_Recognition_in_the_CVPR_2021_paper.pdf)\n- Improving Transferability of Adversarial Patches on Face Recognition With Generative Models [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FXiao_Improving_Transferability_of_Adversarial_Patches_on_Face_Recognition_With_Generative_CVPR_2021_paper.pdf)\n- **WebFace260M**: A Benchmark Unveiling the Power of Million-Scale Deep Face Recognition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhu_WebFace260M_A_Benchmark_Unveiling_the_Power_of_Million-Scale_Deep_Face_CVPR_2021_paper.pdf)\n- **FaceSec**: A Fine-Grained Robustness Evaluation Framework for Face Recognition Systems [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FTong_FaceSec_A_Fine-Grained_Robustness_Evaluation_Framework_for_Face_Recognition_Systems_CVPR_2021_paper.pdf)\n- Spherical Confidence Learning for Face Recognition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Spherical_Confidence_Learning_for_Face_Recognition_CVPR_2021_paper.pdf)\n- Consistent Instance False Positive Improves Fairness in Face Recognition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FXu_Consistent_Instance_False_Positive_Improves_Fairness_in_Face_Recognition_CVPR_2021_paper.pdf)\n- **CRFace**: Confidence Ranker for Model-Agnostic Face Detection Refinement [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FVesdapunt_CRFace_Confidence_Ranker_for_Model-Agnostic_Face_Detection_Refinement_CVPR_2021_paper.pdf)\n- **HLA-Face**: Joint High-Low Adaptation for Low Light Face Detection [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FWang_HLA-Face_Joint_High-Low_Adaptation_for_Low_Light_Face_Detection_CVPR_2021_paper.pdf)\n- Structure-Aware Face Clustering on a Large-Scale Graph With 107 Nodes [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FShen_Structure-Aware_Face_Clustering_on_a_Large-Scale_Graph_With_107_Nodes_CVPR_2021_paper.pdf)\n- **img2pose**: Face Alignment and Detection via 6DoF, Face Pose Estimation [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FAlbiero_img2pose_Face_Alignment_and_Detection_via_6DoF_Face_Pose_Estimation_CVPR_2021_paper.pdf)\n- **Clusformer**: A Transformer Based Clustering Approach to Unsupervised Large-Scale Face and Visual Landmark Recognition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FNguyen_Clusformer_A_Transformer_Based_Clustering_Approach_to_Unsupervised_Large-Scale_Face_CVPR_2021_paper.pdf)\n- Continuous Face Aging via Self-Estimated Residual Age Embedding [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Continuous_Face_Aging_via_Self-Estimated_Residual_Age_Embedding_CVPR_2021_paper.pdf)\n- When Age-Invariant Face Recognition Meets Face Age Synthesis: A Multi-Task Learning Framework [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FHuang_When_Age-Invariant_Face_Recognition_Meets_Face_Age_Synthesis_A_Multi-Task_CVPR_2021_paper.pdf)\n- **SDD-FIQA**: Unsupervised Face Image Quality Assessment With Similarity Distribution Distance [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FOu_SDD-FIQA_Unsupervised_Face_Image_Quality_Assessment_With_Similarity_Distribution_Distance_CVPR_2021_paper.pdf)\n- **TediGAN**: Text-Guided Diverse Face Image Generation and Manipulation [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FXia_TediGAN_Text-Guided_Diverse_Face_Image_Generation_and_Manipulation_CVPR_2021_paper.pdf)\n- GAN Prior Embedded Network for Blind Face Restoration in the Wild [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FYang_GAN_Prior_Embedded_Network_for_Blind_Face_Restoration_in_the_CVPR_2021_paper.pdf)\n- Inverting Generative Adversarial Renderer for Face Reconstruction [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FPiao_Inverting_Generative_Adversarial_Renderer_for_Face_Reconstruction_CVPR_2021_paper.pdf)\n- Progressive Semantic-Aware Style Transformation for Blind Face Restoration [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FChen_Progressive_Semantic-Aware_Style_Transformation_for_Blind_Face_Restoration_CVPR_2021_paper.pdf)\n- Towards Real-World Blind Face Restoration With Generative Facial Prior [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FWang_Towards_Real-World_Blind_Face_Restoration_With_Generative_Facial_Prior_CVPR_2021_paper.pdf)\n- **FaceInpainter**: High Fidelity Face Adaptation to Heterogeneous Domains [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_FaceInpainter_High_Fidelity_Face_Adaptation_to_Heterogeneous_Domains_CVPR_2021_paper.pdf)\n- One Shot Face Swapping on Megapixels [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhu_One_Shot_Face_Swapping_on_Megapixels_CVPR_2021_paper.pdf)\n- High-Fidelity and Arbitrary Face Editing [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FGao_High-Fidelity_and_Arbitrary_Face_Editing_CVPR_2021_paper.pdf)\n- Seeking the Shape of Sound: An Adaptive Framework for Learning Voice-Face Association [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FWen_Seeking_the_Shape_of_Sound_An_Adaptive_Framework_for_Learning_CVPR_2021_paper.pdf)\n- Flow-Guided One-Shot Talking Face Generation With a High-Resolution Audio-Visual Dataset [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhang_Flow-Guided_One-Shot_Talking_Face_Generation_With_a_High-Resolution_Audio-Visual_Dataset_CVPR_2021_paper.pdf)\n- Pose-Controllable Talking Face Generation by Implicitly Modularized Audio-Visual Representation [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhou_Pose-Controllable_Talking_Face_Generation_by_Implicitly_Modularized_Audio-Visual_Representation_CVPR_2021_paper.pdf)\n- Monocular Reconstruction of Neural Face Reflectance Fields [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FR_Monocular_Reconstruction_of_Neural_Face_Reflectance_Fields_CVPR_2021_paper.pdf)\n- Learning Complete 3D Morphable Face Models From Images and Videos [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FR_Learning_Complete_3D_Morphable_Face_Models_From_Images_and_Videos_CVPR_2021_paper.pdf)\n- Riggable 3D Face Reconstruction via In-Network Optimization [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FBai_Riggable_3D_Face_Reconstruction_via_In-Network_Optimization_CVPR_2021_paper.pdf)\n- Learning To Aggregate and Personalize 3D Face From In-the-Wild Photo Collection [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhang_Learning_To_Aggregate_and_Personalize_3D_Face_From_In-the-Wild_Photo_CVPR_2021_paper.pdf)\n- Lifting 2D StyleGAN for 3D-Aware Face Generation [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FShi_Lifting_2D_StyleGAN_for_3D-Aware_Face_Generation_CVPR_2021_paper.pdf)\n- **3DCaricShop**: A Dataset and a Baseline Method for Single-View 3D Caricature Face Reconstruction [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FQiu_3DCaricShop_A_Dataset_and_a_Baseline_Method_for_Single-View_3D_CVPR_2021_paper.pdf)\n- Pareidolia Face Reenactment [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FSong_Pareidolia_Face_Reenactment_CVPR_2021_paper.pdf)\n- Lips Don't Lie: A Generalisable and Robust Approach To Face Forgery Detection [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FHaliassos_Lips_Dont_Lie_A_Generalisable_and_Robust_Approach_To_Face_CVPR_2021_paper.pdf)\n- Spatial-Phase Shallow Learning: Rethinking Face Forgery Detection in Frequency Domain [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLiu_Spatial-Phase_Shallow_Learning_Rethinking_Face_Forgery_Detection_in_Frequency_Domain_CVPR_2021_paper.pdf)\n- Frequency-Aware Discriminative Feature Learning Supervised by Single-Center Loss for Face Forgery Detection [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Frequency-Aware_Discriminative_Feature_Learning_Supervised_by_Single-Center_Loss_for_Face_CVPR_2021_paper.pdf)\n- Generalizing Face Forgery Detection With High-Frequency Features [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLuo_Generalizing_Face_Forgery_Detection_With_High-Frequency_Features_CVPR_2021_paper.pdf)\n- Face Forgery Detection by 3D Decomposition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhu_Face_Forgery_Detection_by_3D_Decomposition_CVPR_2021_paper.pdf)\n- Exploring Adversarial Fake Images on Face Manifold [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Exploring_Adversarial_Fake_Images_on_Face_Manifold_CVPR_2021_paper.pdf)\n- Representative Forgery Mining for Fake Face Detection [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FWang_Representative_Forgery_Mining_for_Fake_Face_Detection_CVPR_2021_paper.pdf)\n- Cross Modal Focal Loss for RGBD Face Anti-Spoofing [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FGeorge_Cross_Modal_Focal_Loss_for_RGBD_Face_Anti-Spoofing_CVPR_2021_paper.pdf)\n- High-Fidelity Face Tracking for AR\u002FVR via Deep Lighting Adaptation [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FChen_High-Fidelity_Face_Tracking_for_ARVR_via_Deep_Lighting_Adaptation_CVPR_2021_paper.pdf)\n- Towards High Fidelity Face Relighting With Realistic Shadows [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FHou_Towards_High_Fidelity_Face_Relighting_With_Realistic_Shadows_CVPR_2021_paper.pdf)\n- **IronMask**: Modular Architecture for Protecting Deep Face Template [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FKim_IronMask_Modular_Architecture_for_Protecting_Deep_Face_Template_CVPR_2021_paper.pdf)\n- Face Forensics in the Wild [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhou_Face_Forensics_in_the_Wild_CVPR_2021_paper.pdf)\n\n**ICCV2021**  \n- Body-Face Joint Detection via Embedding and Head Hook [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FWan_Body-Face_Joint_Detection_via_Embedding_and_Head_Hook_ICCV_2021_paper.pdf)\n- Adaptive Label Noise Cleaning With Meta-Supervision for Deep Face Recognition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FZhang_Adaptive_Label_Noise_Cleaning_With_Meta-Supervision_for_Deep_Face_Recognition_ICCV_2021_paper.pdf)  \n- Teacher-Student Adversarial Depth Hallucination To Improve Face Recognition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FUppal_Teacher-Student_Adversarial_Depth_Hallucination_To_Improve_Face_Recognition_ICCV_2021_paper.pdf)  \n- **DAM**: Discrepancy Alignment Metric for Face Recognition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FLiu_DAM_Discrepancy_Alignment_Metric_for_Face_Recognition_ICCV_2021_paper.pdf)\n- **SynFace**: Face Recognition With Synthetic Data [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FQiu_SynFace_Face_Recognition_With_Synthetic_Data_ICCV_2021_paper.pdf)\n- **PASS**: Protected Attribute Suppression System for Mitigating Bias in Face Recognition [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FDhar_PASS_Protected_Attribute_Suppression_System_for_Mitigating_Bias_in_Face_ICCV_2021_paper.pdf)\n- Disentangled Representation for Age-Invariant Face Recognition: A Mutual Information Minimization Perspective [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FHou_Disentangled_Representation_for_Age-Invariant_Face_Recognition_A_Mutual_Information_Minimization_ICCV_2021_paper.pdf)  \n- Learning Facial Representations From the Cycle-Consistency of Face [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FChang_Learning_Facial_Representations_From_the_Cycle-Consistency_of_Face_ICCV_2021_paper.pdf)  \n- Personalized and Invertible Face De-Identification by Disentangled Identity Information Manipulation [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FCao_Personalized_and_Invertible_Face_De-Identification_by_Disentangled_Identity_Information_Manipulation_ICCV_2021_paper.pdf)  \n- **ADNet**: Leveraging Error-Bias Towards Normal Direction in Face Alignment [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FHuang_ADNet_Leveraging_Error-Bias_Towards_Normal_Direction_in_Face_Alignment_ICCV_2021_paper.pdf)\n- Towards Face Encryption by Generating Adversarial Identity Masks [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FYang_Towards_Face_Encryption_by_Generating_Adversarial_Identity_Masks_ICCV_2021_paper.pdf)  \n- A Latent Transformer for Disentangled Face Editing in Images and Videos [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FYao_A_Latent_Transformer_for_Disentangled_Face_Editing_in_Images_and_ICCV_2021_paper.pdf)  \n- **Re-Aging GAN**: Toward Personalized Face Age Transformation [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FMakhmudkhujaev_Re-Aging_GAN_Toward_Personalized_Face_Age_Transformation_ICCV_2021_paper.pdf)\n- Disentangled Lifespan Face Synthesis [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FHe_Disentangled_Lifespan_Face_Synthesis_ICCV_2021_paper.pdf)\n- Face Image Retrieval With Attribute Manipulation [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FZaeemzadeh_Face_Image_Retrieval_With_Attribute_Manipulation_ICCV_2021_paper.pdf)\n- Self-Supervised 3D Face Reconstruction via Conditional Estimation [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FWen_Self-Supervised_3D_Face_Reconstruction_via_Conditional_Estimation_ICCV_2021_paper.pdf)  \n- Towards High Fidelity Monocular Face Reconstruction With Rich Reflectance Using Self-Supervised Learning and Ray Tracing [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FDib_Towards_High_Fidelity_Monocular_Face_Reconstruction_With_Rich_Reflectance_Using_ICCV_2021_paper.pdf)\n- Self-Supervised 3D Face Reconstruction via Conditional Estimation [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FWen_Self-Supervised_3D_Face_Reconstruction_via_Conditional_Estimation_ICCV_2021_paper.pdf)  \n- Topologically Consistent Multi-View Face Inference Using Volumetric Sampling [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FLi_Topologically_Consistent_Multi-View_Face_Inference_Using_Volumetric_Sampling_ICCV_2021_paper.pdf)\n- **Fake It Till You Make It**: Face Analysis in the Wild Using Synthetic Data Alone [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FWood_Fake_It_Till_You_Make_It_Face_Analysis_in_the_ICCV_2021_paper.pdf)\n- Exploring Temporal Coherence for More General Video Face Forgery Detection [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FZheng_Exploring_Temporal_Coherence_for_More_General_Video_Face_Forgery_Detection_ICCV_2021_paper.pdf)  \n- **OpenForensics**: Large-Scale Challenging Dataset for Multi-Face Forgery Detection and Segmentation In-the-Wild [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FLe_OpenForensics_Large-Scale_Challenging_Dataset_for_Multi-Face_Forgery_Detection_and_Segmentation_ICCV_2021_paper.pdf)\n- Detection and Continual Learning of Novel Face Presentation Attacks [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FRostami_Detection_and_Continual_Learning_of_Novel_Face_Presentation_Attacks_ICCV_2021_paper.pdf)\n- **MeshTalk**: 3D Face Animation From Speech Using Cross-Modality Disentanglement [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FRichard_MeshTalk_3D_Face_Animation_From_Speech_Using_Cross-Modality_Disentanglement_ICCV_2021_paper.pdf)  \n- Super-Resolving Cross-Domain Face Miniatures by Peeking at One-Shot Exemplar [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FLi_Super-Resolving_Cross-Domain_Face_Miniatures_by_Peeking_at_One-Shot_Exemplar_ICCV_2021_paper.pdf)  \n- Multi-Modality Associative Bridging Through Memory: Speech Sound Recollected From Face Video [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FKim_Multi-Modality_Associative_Bridging_Through_Memory_Speech_Sound_Recollected_From_Face_ICCV_2021_paper.pdf)  \n- **FACIAL**: Synthesizing Dynamic Talking Face With Implicit Attribute Learning [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FZhang_FACIAL_Synthesizing_Dynamic_Talking_Face_With_Implicit_Attribute_Learning_ICCV_2021_paper.pdf)  \n- **VariTex**: Variational Neural Face Textures [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FBuhler_VariTex_Variational_Neural_Face_Textures_ICCV_2021_paper.pdf)  \n- Learning High-Fidelity Face Texture Completion Without Complete Face Texture [[paper]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FKim_Learning_High-Fidelity_Face_Texture_Completion_Without_Complete_Face_Texture_ICCV_2021_paper.pdf)  \n\n###### 2020-10-02\n**ICIP2020**  \n- 3D SPARSE DEFORMATION SIGNATURE FOR DYNAMIC FACE RECOGNITION\n- A Stacking Ensemble for Anomaly Based Client-Specific Face Spoofing Detection\n- ADAPTIVE AGGREGATED TRACKLET LINKING FOR MULTI-FACE TRACKING\n- ATTENTION SELECTIVE NETWORK FOR FACE SYNTHESIS AND POSE-INVARIANT FACE RECOGNITION\n- **EDGE-GAN**: EDGE CONDITIONED MULTI-VIEW FACE IMAGE GENERATION\n- EXTRACTING DEEP LOCAL FEATURES TO DETECT MANIPULATED IMAGES OF HUMAN FACES\n- FACE AUTHENTICATION FROM GRAYSCALE CODED LIGHT FIELD\n- FACE RECOGNITION UNDER LOW ILLUMINATION VIA DEEP FEATURE RECONSTRUCTION NETWORK\n- IMPROVING DETECTION AND RECOGNITION OF DEGRADED FACES BY DISCRIMINATIVE FEATURE RESTORATION USING GAN\n- **QAMFACE**: QUADRATIC ADDITIVE ANGULAR MARGIN LOSS FOR FACE RECOGNITION\n- REALISTIC TALKING FACE SYNTHESIS WITH GEOMETRY-AWARE FEATURE TRANSFORMATION\n- TRIPLET DISTILLATION FOR DEEP FACE RECOGNITION  \n\n**ECCV2020**\n- **“Look Ma, no landmarks!”** – Unsupervised, Model-based Dense Face Alignment\n- Hierarchical Face Aging through Disentangled Latent Characteristics\n- Semi-Siamese Training for Shallow Face Learning\n- Face Super-Resolution Guided by 3D Facial Priors\n- Personalized Face Modeling for Improved Face Reconstruction and Motion Retargeting\n- **ProgressFace**: Scale-Aware Progressive Learning for Face Detection\n- Face Anti-Spoofing with Human Material Perception\n- **Beyond 3DMM Space**: Towards Fine-grained 3D Face Reconstruction\n- Blind Face Restoration via Deep Multi-scale Component Dictionaries\n- Inequality-Constrained and Robust 3D Face Model Fitting\n- **BroadFace**: Looking at Tens of Thousands of People at Once for Face Recognition\n- Explainable Face Recognition\n- **CONFIG**: Controllable Neural Face Image Generation\n- **Sub-center ArcFace**: Boosting Face Recognition by Large-scale Noisy Web Faces\n- **CelebA-Spoof**: Large-Scale Face Anti-Spoofing Dataset with Rich Annotations\n- **Thinking in Frequency**: Face Forgery Detection by Mining Frequency-aware Clues\n- Edge-aware Graph Representation Learning and Reasoning for Face Parsing\n- Learning Flow-based Feature Warping for Face Frontalization with Illumination Inconsistent Supervision\n- **CAFE-GAN**: Arbitrary Face Attribute Editing with Complementary Attention Feature\n- Self-Supervised Monocular 3D Face Reconstruction by Occlusion-Aware Multi-view Geometry Consistency\n- **Generate to Adapt**: Resolution Adaption Network for Surveillance Face Recognition\n- **Caption-Supervised Face Recognition**: Training a State-of-the-Art Face Model without Manual Annotation\n- Design and Interpretation of Universal Adversarial Patches in Face Detection\n- **JNR**: Joint-based Neural Rig Representation for Compact 3D Face Modeling\n- On Disentangling Spoof Trace for Generic Face Anti-Spoofing\n- Towards causal benchmarking of bias in face analysis algorithms\n- Towards Fast, Accurate and Stable 3D Dense Face Alignment\n- Face Anti-Spoofing via Disentangled Representation Learning\n- Learning to Predict Salient Faces: A Novel Visual-Audio Saliency Model\n- **MEAD**: A Large-scale Audio-visual Dataset for Emotional Talking-face Generation\n- **Margin-Mix**: Semi–Supervised Learning for Face Expression Recognition\n- Password-conditioned Anonymization and Deanonymization with Face Identity Transformers\n- Improving Face Recognition by Clustering Unlabeled Faces in the Wild\n- Exclusivity-Consistency Regularized Knowledge Distillation for Face Recognition\n- **BioMetricNet**: deep unconstrained face verification through learning of metrics regularized onto Gaussian distributions\n- Eyeglasses 3D shape reconstruction from a single face image\n- Deep Cross-species Feature Learning for Animal Face Recognition via Residual Interspecies Equivariant Network\n- High Resolution Zero-Shot Domain Adaptation of Synthetically Rendered Face Images\n- **ByeGlassesGAN**: Identity Preserving Eyeglasses Removal for Face Images\n- Jointly De-biasing Face Recognition and Demographic Attribute Estimation\n- Synthesizing Coupled 3D Face Modalities by Trunk-Branch Generative Adversarial Networks\n- Improving Face Recognition from Hard Samples via Distribution Distillation Loss\n- Manifold Projection for Adversarial Defense on Face Recognition  \n\n**SIGGRAPH2020**\n- A System for Efficient 3D Printed Stop-Motion Face Animation\n- Accurate Face Rig Approximation With Deep Differential Subspace Reconstruction\n- **DeepFaceDrawing**: Deep Generation of Face Images from Sketches\n- **The Eyes Have It**: An Integrated Eye and Face Model for Photorealistic Facial Animation  \n\n**IJCAI2020**  \n- Biased Feature Learning for Occlusion Invariant Face Recognition\n- Reference Guided Face Component Editing\n- Arbitrary Talking Face Generation via Attentional Audio-Visual Coherence Learning\n- **FakeSpotter**: A Simple yet Robust Baseline for Spotting AI-Synthesized Fake Faces  \n\n**CVPR2020**  \n- Cross-Modal Deep Face Normals With Deactivable Skip Connections\n- One-Shot Domain Adaptation for Face Generation\n- Towards Learning Structure via Consensus for Face Segmentation and Parsing\n- **BFBox**: Searching Face-Appropriate Backbone and Feature Pyramid Network for Face Detector\n- **Domain Balancing**: Face Recognition on Long-Tailed Domains\n- **FReeNet**: Multi-Identity Face Reenactment\n- Learning Identity-Invariant Motion Representations for Cross-ID Face Reenactment\n- **Global-Local GCN**: Large-Scale Label Noise Cleansing for Face Recognition\n- **3FabRec**: Fast Few-Shot Face Alignment by Reconstruction\n- Global Texture Enhancement for Fake Face Detection in the Wild\n- **CurricularFace**: Adaptive Curriculum Learning Loss for Deep Face Recognition\n- On the Detection of Digital Face Manipulation\n- Deep Spatial Gradient and Temporal Depth Learning for Face Anti-Spoofing\n- **ReDA**:Reinforced Differentiable Attribute for 3D Face Reconstruction\n- Cross-Domain Face Presentation Attack Detection via Multi-Domain Disentangled Representation Learning\n- **FaceScape**: A Large-Scale High Quality 3D Face Dataset and Detailed Riggable 3D Face Prediction\n- Interpreting the Latent Space of GANs for Semantic Face Editing\n- **Rotate-and-Render**: Unsupervised Photorealistic Face Rotation From Single-View Images\n- Disentangled and Controllable Face Image Generation via 3D Imitative-Contrastive Learning\n- Density-Aware Feature Embedding for Face Clustering\n- Learning to Have an Ear for Face Super-Resolution\n- Learning Formation of Physically-Based Face Attributes\n- **LUVLi Face Alignment**: Estimating Landmarks' Location, Uncertainty, and Visibility Likelihood\n- Learning Meta Face Recognition in Unseen Domains\n- Cross-Spectral Face Hallucination via Disentangling Independent Factors\n- Deep Face Super-Resolution With Iterative Collaboration Between Attentive Recovery and Landmark Estimation\n- Data Uncertainty Learning in Face Recognition\n- Face X-Ray for More General Face Forgery Detection\n- **Vec2Face**: Unveil Human Faces From Their Blackbox Features in Face Recognition\n- **FM2u-Net**: Face Morphological Multi-Branch Network for Makeup-Invariant Face Verification\n- Uncertainty-Aware Mesh Decoder for High Fidelity 3D Face Reconstruction\n- Enhanced Blind Face Restoration With Multi-Exemplar Images and Adaptive Spatial Feature Fusion\n- **SER-FIQ**: Unsupervised Estimation of Face Image Quality Based on Stochastic Embedding Robustness\n- Towards High-Fidelity 3D Face Reconstruction From In-the-Wild Images Using Graph Convolutional Networks\n- **RDCFace**: Radial Distortion Correction for Face Recognition\n- Searching Central Difference Convolutional Networks for Face Anti-Spoofing\n- **RetinaFace**: Single-Shot Multi-Level Face Localisation in the Wild\n- Mitigating Bias in Face Recognition Using Skewness-Aware Reinforcement Learning\n- **DeeperForensics-1.0**: A Large-Scale Dataset for Real-World Face Forgery Detection\n- **GroupFace**: Learning Latent Groups and Constructing Group-Based Representations for Face Recognition\n- A Morphable Face Albedo Model\n- Learning Oracle Attention for High-Fidelity Face Completion\n- Learning Physics-Guided Face Relighting Under Directional Light\n- Towards Universal Representation Learning for Deep Face Recognition\n- Rotation Consistent Margin Loss for Efficient Low-Bit Face Recognition\n- **HAMBox**: Delving Into Mining High-Quality Anchors on Face Detection\n- Hierarchical Pyramid Diverse Attention Networks for Face Recognition\n- Dynamic Face Video Segmentation via Reinforcement Learning\n- Copy and Paste GAN: Face Hallucination From Shaded Thumbnails\n- Single-Side Domain Generalization for Face Anti-Spoofing  \n\n**AAAI2020** \n- Fast and Robust Face-to-Parameter Translation for Game Character Auto-Creation\n- Mis-classified Vector Guided Softmax Loss for Face Recognition\n- Learning Meta Model for Zero- and Few-shot Face Anti-spoofing\n- Learning to Deblur Face Images via Sketch Synthesis\n- **FAN-Face**: a simple orthogonal improvement to deep face recognition\n- **KPNet**: Towards Minimal Face Detector\n- Towards Omni-Supervised Face Alignment for Large Scale Unlabeled Videos\n- Regularized Fine-grained Meta Face Anti-spoofing\n- **GDFace**: Gated Deformation for Multi-view Face Image Synthesis\n- Realistic Face Reenactment via Self-Supervised Disentangling of Identity and Pose\n- **MarioNETte**: Few-shot Face Reenactment Preserving Identity of Unseen Targets\n- A New Dataset and Boundary-Attention Semantic Segmentation for Face Parsing\n- Video Face Super-Resolution with Motion-Adaptive Feedback Cell\n- Facial Attribute Capsules for Noise Face Super Resolution\n- Joint Super-Resolution and Alignment of Tiny Faces\n###### 2020-01-26\n- **UGG**: Uncertainty Modeling of Contextual-Connections Between Tracklets for Unconstrained Video-Based Face Recognition\n- **PDSN**: Occlusion Robust Face Recognition Based on Mask Learning With Pairwise Differential Siamese Network\n- Attentional Feature-Pair Relation Networks for Accurate Face Recognition\n- **PFE**: Probabilistic Face Embeddings\n- Towards Interpretable Face Recognition\n- **Co-Mining**: Deep Face Recognition With Noisy Labels\n- **Fair Loss**: Margin-Aware Reinforcement Learning for Deep Face Recognition\n- Discriminatively Learned Convex Models for Set Based Face Recognition\n- **DVG**: Dual Variational Generation for Low Shot Heterogeneous Face Recognition\n- **CDP**: Consensus-Driven Propagation in Massive Unlabeled Data for Face Recognition   \n\n- **BCL**: Video Face Clustering With Unknown Number of Clusters  \n\n- **DeCaFA**: Deep Convolutional Cascade for Face Alignment in the Wild\n- **AWing**: Adaptive Wing Loss for Robust Face Alignment via Heatmap Regression\n- **KDN**: Face Alignment With Kernel Density Deep Neural Network  \n\n- **DF2Net**: A Dense-Fine-Finer Network for Detailed 3D Face Reconstruction\n- Face Video Deblurring Using 3D Facial Priors\n- Semi-Supervised Monocular 3D Face Reconstruction With End-to-End Shape-Preserved Domain Transfer\n- **3DFC**: 3D Face Modeling From Diverse Raw Scan Data  \n\n- Live Face De-Identification in Video\n- Face-to-Parameter Translation for Game Character Auto-Creation\n- **SC-FEGAN**: Face Editing Generative Adversarial Network With User's Sketch and Color\n- **FSGAN**: Subject Agnostic Face Swapping and Reenactment\n- **Make a Face**: Towards Arbitrary High Fidelity Face Manipulation\n- Face De-Occlusion Using 3D Morphable Model and Generative Adversarial Network\n- **FRV**: Face Reconstruction from Voice using Generative Adversarial Networks\n- **From Inference to Generation**: End-to-end Fully Self-supervised Generation of Human Face from Speech\n- **PFSR**: Progressive Face Super-Resolution via Attention to Facial Landmark\n\n###### 2019-07-11\n- Deep face recognition using imperfect facial data\n- Unequal-Training for Deep Face Recognition With Long-Tailed Noisy Data\n- **RegularFace**: Deep Face Recognition via Exclusive Regularization\n- **UniformFace**: Learning Deep Equidistributed Representation for Face Recognition\n- **P2SGrad**: Refined Gradients for Optimizing Deep Face Models\n- **AdaptiveFace**: Adaptive Margin and Sampling for Face Recognition\n- **AdaCos**: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations\n- Low-Rank Laplacian-Uniform Mixed Model for Robust Face Recognition\n- **NoiseFace**: Noise-Tolerant Paradigm for Training Face Recognition CNNs\n- Feature Transfer Learning for Face Recognition With Under-Represented Data\n- **Led3D**: A Lightweight and Efficient Deep Approach to Recognizing Low-Quality 3D Faces\n- R3 Adversarial Network for Cross Model Face Recognition\n\n- **RetinaFace**: Single-stage Dense Face Localisation in the Wild\n- Group Sampling for Scale Invariant Face Detection\n- **FA-RPN**: Floating Region Proposals for Face Detection\n\n- **Semantic Alignment**: Finding Semantically Consistent Ground-Truth for Facial Landmark Detection\n- Robust Facial Landmark Detection via Occlusion-Adaptive Deep Networks\n\n- **LTC**: Learning to Cluster Faces on an Affinity Graph\n\n- **FECNet**: A Compact Embedding for Facial Expression Similarity\n- **LBVCNN**: Local Binary Volume Convolutional Neural Network for Facial Expression Recognition from Image Sequences\n\n- Joint Representation and Estimator Learning for Facial Action Unit Intensity Estimation\n- Local Relationship Learning With Person-Specific Shape Regularization for Facial Action Unit Detection\n- **TCAE**: Self-Supervised Representation Learning From Videos for Facial Action Unit Detection\n- **JAANet**: Deep Adaptive Attention for Joint Facial Action Unit Detection and Face Alignment\n\n- **2DASL**: Joint 3D Face Reconstruction and Dense Face Alignment from A Single Image with 2D-Assisted Self-Supervised Learning\n- **MVF-Net**: Multi-View 3D Face Morphable Model Regression\n- Dense 3D Face Decoding Over 2500FPS: Joint Texture & Shape Convolutional Mesh Decoders\n- Towards High-Fidelity Nonlinear 3D Face Morphable Model\n- Combining 3D Morphable Models: A Large Scale Face-And-Head Model\n- Disentangled Representation Learning for 3D Face Shap\n- Self-Supervised Adaptation of High-Fidelity Face Models for Monocular Performance Tracking\n- **MMFace**: A Multi-Metric Regression Network for Unconstrained Face Reconstruction\n- Learning to Regress 3D Face Shape and Expression From an Image Without 3D Supervision\n- Boosting Local Shape Matching for Dense 3D Face Correspondence\n- **FML**: Face Model Learning From Videos\n- **2DASL**: Joint 3D Face Reconstruction and Dense Face Alignment from A Single Image with 2D-Assisted Self-Supervised Learning\n\n- **ATVGnet**: Hierarchical Cross-Modal Talking Face Generation With Dynamic Pixel-Wise Loss\n- **Speech2Face**: Learning the Face Behind a Voice\n\n- Unsupervised Face Normalization With Extreme Pose and Expression in the Wild\n- **GANFIT**: Generative Adversarial Network Fitting for High Fidelity 3D Face Reconstruction\n\n- **BeautyGAN**: Instance-level Facial Makeup Transfer with Deep Generative Adversarial Network\n- **FUNIT**: Few-Shot Unsupervised Image-to-Image Translation\n- Automatic Face Aging in Videos via Deep Reinforcement Learning\n- Attribute-Aware Face Aging With Wavelet-Based Generative Adversarial Networks\n- **SAGAN**:Generative Adversarial Network with Spatial Attention for Face Attribute Editing\n- **APDrawingGAN**: Generating Artistic Portrait Drawings From Face Photos With Hierarchical GANs\n- **StyleGAN**: A Style-Based Generator Architecture for Generative Adversarial Networks\n\n- 3D Guided Fine-Grained Face Manipulation\n- **SemanticComponent**: Semantic Component Decomposition for Face Attribute Manipulation\n\n- **Dataset and Benchmark**: A Dataset and Benchmark for Large-Scale Multi-Modal Face Anti-Spoofing\n- Deep Tree Learning for Zero-Shot Face Anti-Spoofing\n\n- Decorrelated Adversarial Learning for Age-Invariant Face Recognition\n- Multi-Adversarial Discriminative Deep Domain Generalization for Face Presentation Attack Detection\n- Efficient Decision-Based Black-Box Adversarial Attacks on Face Recognition \n\n- **Speech2Face**: Learning the Face Behind a Voice\n- **JFDFMR**: Joint Face Detection and Facial Motion Retargeting for Multiple Faces\n- **ATVGnet**: Hierarchical Cross-Modal Talking Face Generation With Dynamic Pixel-Wise Loss\n\n- High-Quality Face Capture Using Anatomical Muscles\n- Monocular Total Capture: Posing Face, Body, and Hands in the Wild\n- Expressive Body Capture: 3D Hands, Face, and Body From a Single Image\n\n###### 2019-04-06\n- **ISRN**: Improved Selective Refinement Network for Face Detection\n- **DSFD**: Dual Shot Face Detector\n- **PyramidBox++**: High Performance Detector for Finding Tiny Face\n- **VIM-FD**: Robust and High Performance Face Detector\n- **SHF**: Robust Face Detection via Learning Small Faces on Hard Images\n- **SRN**: Selective Refinement Network for High Performance Face Detection\n- **SFDet**: Single-Shot Scale-Aware Network for Real-Time Face Detection\n- **JFDFMR**: Joint Face Detection and Facial Motion Retargeting for Multiple Faces\n- **PFLD**: A Practical Facial Landmark Detector\n- **LinkageFace**: Linkage Based Face Clustering via Graph Convolution Network\n- **MLT**: Face Recognition: A Novel Multi-Level Taxonomy based Survey\n- **GhostVLAD**: GhostVLAD for set-based face recognition\n- **DocFace+**: ID Document to Selfie Matching\n- **DiF**: Diversity in Faces\n- **2018Survey**: Face Recognition: From Traditional to Deep Learning Methods\n\n###### 2019-01-12\n- **2018Survey**: Deep Facial Expression Recognition: A Survey\n- **2018Survey**: Deep Face Recognition: A Survey\n- **SphereFace+(MHE)**: Learning towards Minimum Hyperspherical Energy\n- **HyperFace**: A Deep Multi-task Learning Framework for Face Detection, Landmark Localization, Pose Estimation, and Gender Recognition\n\n###### 2018-12-01\n- **FRVT**: Face Recognition Vendor Test\n- **GANimation**: Anatomically-aware Facial Animation from a Single Image\n- **StarGAN**: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation\n- **Faceswap**: A tool that utilizes deep learning to recognize and swap faces in pictures and videos\n- **HF-PIM**: Learning a High Fidelity Pose Invariant Model for High-resolution Face Frontalization\n- **PRNet**: Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Network\n- **LAB**: Look at Boundary: A Boundary-Aware Face Alignment Algorithm\n- **Super-FAN**: Integrated facial landmark localization and super-resolution of real-world low resolution faces in arbitrary poses with GANs\n- **Face-Alignment**: How far are we from solving the 2D & 3D Face Alignment problem? (and a dataset of 230,000 3D facial landmarks)\n- **Face3D**: Python tools for processing 3D face\n- **IMDb-Face**: The Devil of Face Recognition is in the Noise\n- **AAM-Softmax(CCL)**: Face Recognition via Centralized Coordinate Learning\n- **AM-Softmax**: Additive Margin Softmax for Face Verification\n- **FeatureIncay**: Feature Incay for Representation Regularization\n- **NormFace**: L2 hypersphere embedding for face Verification\n- **CocoLoss**: Rethinking Feature Discrimination and Polymerization for Large-scale Recognition\n- **L-Softmax**: Large-Margin Softmax Loss for Convolutional Neural Networks\n\n###### 2018-07-21\n- **MobileFace**: A face recognition solution on mobile device\n- **Trillion Pairs**: Challenge 3: Face Feature Test\u002FTrillion Pairs\n- **MobileFaceNets**: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices\n\n###### 2018-04-20\n- **PyramidBox**: A Context-assisted Single Shot Face Detector\n- **PCN**: Real-Time Rotation-Invariant Face Detection with Progressive Calibration Networks\n- **S³FD**: Single Shot Scale-invariant Face Detector\n- **SSH**: Single Stage Headless Face Detector\n- **NPD**: A Fast and Accurate Unconstrained Face Detector\n- **PICO**: Object Detection with Pixel Intensity Comparisons Organized in Decision Trees\n- **libfacedetection**: A fast binary library for face detection and face landmark detection in images.\n- **SeetaFaceEngine**: SeetaFace Detection, SeetaFace Alignment and SeetaFace Identification.\n- **FaceID**: An implementation of iPhone X's FaceID using face embeddings and siamese networks on RGBD images.\n\n###### 2018-03-28\n- **InsightFace(ArcFace)**: 2D and 3D Face Analysis Project\n- **CosFace**: Large Margin Cosine Loss for Deep Face Recognition\n\n## 🔖 Face Benchmark and Dataset\n#### Face Recognition Data\n- **DiF**: Diversity in Faces [[project]](https:\u002F\u002Fwww.research.ibm.com\u002Fartificial-intelligence\u002Ftrusted-ai\u002Fdiversity-in-faces\u002F) [[blog]](https:\u002F\u002Fwww.ibm.com\u002Fblogs\u002Fresearch\u002F2019\u002F01\u002Fdiversity-in-faces\u002F)\n- **FRVT**: Face Recognition Vendor Test [[project]](https:\u002F\u002Fwww.nist.gov\u002Fprograms-projects\u002Fface-recognition-vendor-test-frvt) [[leaderboard]](https:\u002F\u002Fwww.nist.gov\u002Fprograms-projects\u002Fface-recognition-vendor-test-frvt-ongoing)\n- **IMDb-Face**: The Devil of Face Recognition is in the Noise(**59k people in 1.7M images**) [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ECCV_2018\u002Fpapers\u002FLiren_Chen_The_Devil_of_ECCV_2018_paper.pdf \"ECCV2018\") [[dataset]](https:\u002F\u002Fgithub.com\u002Ffwang91\u002FIMDb-Face)\n- **Trillion Pairs**: Challenge 3: Face Feature Test\u002FTrillion Pairs(**MS-Celeb-1M-v1c with 86,876 ids\u002F3,923,399 aligned images  + Asian-Celeb 93,979 ids\u002F2,830,146 aligned images**) [[benckmark]](http:\u002F\u002Ftrillionpairs.deepglint.com\u002Foverview \"DeepGlint\") [[dataset]](http:\u002F\u002Ftrillionpairs.deepglint.com\u002Fdata) [[result]](http:\u002F\u002Ftrillionpairs.deepglint.com\u002Fresults)\n- **MF2**: Level Playing Field for Million Scale Face Recognition(**672K people in 4.7M images**) [[paper]](https:\u002F\u002Fhomes.cs.washington.edu\u002F~kemelmi\u002Fms.pdf \"CVPR2017\") [[dataset]](http:\u002F\u002Fmegaface.cs.washington.edu\u002Fdataset\u002Fdownload_training.html) [[result]](http:\u002F\u002Fmegaface.cs.washington.edu\u002Fresults\u002Ffacescrub_challenge2.html) [[benckmark]](http:\u002F\u002Fmegaface.cs.washington.edu\u002F)\n- **MegaFace**: The MegaFace Benchmark: 1 Million Faces for Recognition at Scale(**690k people in 1M images**) [[paper]](http:\u002F\u002Fmegaface.cs.washington.edu\u002FKemelmacherMegaFaceCVPR16.pdf \"CVPR2016\") [[dataset]](http:\u002F\u002Fmegaface.cs.washington.edu\u002Fparticipate\u002Fchallenge.html) [[result]](http:\u002F\u002Fmegaface.cs.washington.edu\u002Fresults\u002Ffacescrub.html) [[benckmark]](http:\u002F\u002Fmegaface.cs.washington.edu\u002F)\n- **UMDFaces**: An Annotated Face Dataset for Training Deep Networks(**8k people in 367k images with pose, 21 key-points and gender**) [[paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1611.01484.pdf \"arXiv2016\") [[dataset]](http:\u002F\u002Fwww.umdfaces.io\u002F)\n- **MS-Celeb-1M**: A Dataset and Benchmark for Large Scale Face Recognition(**100K people in 10M images**) [[paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1607.08221.pdf \"ECCV2016\") [[dataset]](http:\u002F\u002Fwww.msceleb.org\u002Fdownload\u002Fsampleset) [[result]](http:\u002F\u002Fwww.msceleb.org\u002Fleaderboard\u002Ficcvworkshop-c1) [[benchmark]](http:\u002F\u002Fwww.msceleb.org\u002F) [[project]](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fproject\u002Fms-celeb-1m-challenge-recognizing-one-million-celebrities-real-world\u002F)\n- **VGGFace2**: A dataset for recognising faces across pose and age(**9k people in 3.3M images**) [[paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.08092.pdf \"arXiv2017\") [[dataset]](http:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vgg\u002Fdata\u002Fvgg_face2\u002F)\n- **VGGFace**: Deep Face Recognition(**2.6k people in 2.6M images**) [[paper]](http:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vgg\u002Fpublications\u002F2015\u002FParkhi15\u002Fparkhi15.pdf \"BMVC2015\") [[dataset]](http:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vgg\u002Fdata\u002Fvgg_face\u002F)\n- **CASIA-WebFace**: Learning Face Representation from Scratch(**10k people in 500k images**) [[paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1411.7923.pdf \"arXiv2014\") [[dataset]](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fenglish\u002FCASIA-WebFace-Database.html)\n- **LFW**: Labeled Faces in the Wild: A Database for Studying Face Recognition in Unconstrained Environments(**5.7k people in 13k images**) [[report]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Flfw\u002Flfw.pdf \"UMASS2007\") [[dataset]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Flfw\u002F#download) [[result]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Flfw\u002Fresults.html) [[benchmark]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Flfw\u002F)\n\n#### Face Detection Data\n- **WiderFace**: WIDER FACE: A Face Detection Benchmark(**400k people in 32k images with a high degree of variability in scale, pose and occlusion**) [[paper]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2016\u002Fpapers\u002FYang_WIDER_FACE_A_CVPR_2016_paper.pdf \"CVPR2016\") [[dataset]](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FWIDERFace\u002F) [[result]](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FWIDERFace\u002FWiderFace_Results.html) [[benchmark]](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FWIDERFace\u002F)\n- **FDDB**: A Benchmark for Face Detection in Unconstrained Settings(**5k faces in 2.8k images**) [[report]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~elm\u002Fpapers\u002Ffddb.pdf \"UMASS2010\") [[dataset]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Ffddb\u002Findex.html#download) [[result]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Ffddb\u002Fresults.html) [[benchmark]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Ffddb\u002F) \n\n#### Face Landmark Data\n- **LS3D-W**: A large-scale 3D face alignment dataset constructed by annotating the images from AFLW, 300VW, 300W and FDDB in a consistent manner with 68 points using the automatic method [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2017\u002Fpapers\u002FBulat_How_Far_Are_ICCV_2017_paper.pdf \"ICCV2017\") [[dataset]](https:\u002F\u002Fadrianbulat.com\u002Fface-alignment)\n- **AFLW**: Annotated Facial Landmarks in the Wild: A Large-scale, Real-world Database for Facial Landmark Localization(**25k faces with 21 landmarks**) [[paper]](https:\u002F\u002Ffiles.icg.tugraz.at\u002Fseafhttp\u002Ffiles\u002F460c7623-c919-4d35-b24e-6abaeacb6f31\u002Fkoestinger_befit_11.pdf \"BeFIT2011\") [[benchmark]](https:\u002F\u002Fwww.tugraz.at\u002Finstitute\u002Ficg\u002Fresearch\u002Fteam-bischof\u002Flrs\u002Fdownloads\u002Faflw\u002F)\n\n#### Face Attribute Data\n- **CelebA**: Deep Learning Face Attributes in the Wild(**10k people in 202k images with 5 landmarks and 40 binary attributes per image**) [[paper]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_iccv_2015\u002Fpapers\u002FLiu_Deep_Learning_Face_ICCV_2015_paper.pdf \"ICCV2015\") [[dataset]](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FCelebA.html)\n\n## 🔖 Face Recognition\n- Live Face De-Identification in Video [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1911.08348 \"ICCV2019\")\n- **UGG**: Uncertainty Modeling of Contextual-Connections Between Tracklets for Unconstrained Video-Based Face Recognition [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.02756 \"ICCV2019\")\n- **PDSN**: Occlusion Robust Face Recognition Based on Mask Learning With Pairwise Differential Siamese Network [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.06290 \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002FlinserSnow\u002FPDSN \"PyTorch\")\n- Attentional Feature-Pair Relation Networks for Accurate Face Recognition [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.06255 \"ICCV2019\")\n- Probabilistic Face Embeddings [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.09658 \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002FseasonSH\u002FProbabilistic-Face-Embeddings \"TensorFlow\") \n- Towards Interpretable Face Recognition [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1805.00611 \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fyubangji123\u002FInterpret_FR \"TensorFlow\") [[project]](http:\u002F\u002Fcvlab.cse.msu.edu\u002Fproject-interpret-FR)\n- **Co-Mining**: Deep Face Recognition With Noisy Labels [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FWang_Co-Mining_Deep_Face_Recognition_With_Noisy_Labels_ICCV_2019_paper.pdf \"ICCV2019\")\n- **Fair Loss**: Margin-Aware Reinforcement Learning for Deep Face Recognition [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FLiu_Fair_Loss_Margin-Aware_Reinforcement_Learning_for_Deep_Face_Recognition_ICCV_2019_paper.pdf \"ICCV2019\")\n- Discriminatively Learned Convex Models for Set Based Face Recognition [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FCevikalp_Discriminatively_Learned_Convex_Models_for_Set_Based_Face_Recognition_ICCV_2019_paper.pdf \"ICCV2019\")\n- **DVG**: Dual Variational Generation for Low Shot Heterogeneous Face Recognition [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.10203 \"NeurIPS2019\") [[code]](https:\u002F\u002Fgithub.com\u002FBradyFU\u002FDVG \"PyTorch\")\n- Deep face recognition using imperfect facial data [[paper]](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS0167739X18331133 \"FGCS2019\")\n- Unequal-Training for Deep Face Recognition With Long-Tailed Noisy Data [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhong_Unequal-Training_for_Deep_Face_Recognition_With_Long-Tailed_Noisy_Data_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fzhongyy\u002FUnequal-Training-for-Deep-Face-Recognition-with-Long-Tailed-Noisy-Data \"MXNet\")\n- **RegularFace**: Deep Face Recognition via Exclusive Regularization [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhao_RegularFace_Deep_Face_Recognition_via_Exclusive_Regularization_CVPR_2019_paper.pdf \"CVPR2019\")\n- **UniformFace**: Learning Deep Equidistributed Representation for Face Recognition [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FDuan_UniformFace_Learning_Deep_Equidistributed_Representation_for_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\")\n- **P2SGrad**: Refined Gradients for Optimizing Deep Face Models [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhang_P2SGrad_Refined_Gradients_for_Optimizing_Deep_Face_Models_CVPR_2019_paper.pdf \"CVPR2019\")\n- **AdaptiveFace**: Adaptive Margin and Sampling for Face Recognition [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FLiu_AdaptiveFace_Adaptive_Margin_and_Sampling_for_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\")\n- **AdaCos**: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhang_AdaCos_Adaptively_Scaling_Cosine_Logits_for_Effectively_Learning_Deep_Face_CVPR_2019_paper.pdf \"CVPR2019\") [[code1]](https:\u002F\u002Fgithub.com\u002Fxialuxi\u002Farcface-caffe \"Caffe\") [[code2]](https:\u002F\u002Fgithub.com\u002F4uiiurz1\u002Fpytorch-adacos \"PyTorch\")\n- Low-Rank Laplacian-Uniform Mixed Model for Robust Face Recognition [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FDong_Low-Rank_Laplacian-Uniform_Mixed_Model_for_Robust_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\")\n- **NoiseFace**: Noise-Tolerant Paradigm for Training Face Recognition CNNs [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FHu_Noise-Tolerant_Paradigm_for_Training_Face_Recognition_CNNs_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fhuangyangyu\u002FNoiseFace \"Caffe\")\n- Feature Transfer Learning for Face Recognition With Under-Represented Data [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FYin_Feature_Transfer_Learning_for_Face_Recognition_With_Under-Represented_Data_CVPR_2019_paper.pdf \"CVPR2019\")\n- **Led3D**: A Lightweight and Efficient Deep Approach to Recognizing Low-Quality 3D Faces [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FMu_Led3D_A_Lightweight_and_Efficient_Deep_Approach_to_Recognizing_Low-Quality_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fmuyouhang\u002FLed3D \"NULL\") [[dataset]](http:\u002F\u002Firip.buaa.edu.cn\u002Flock3dface\u002Findex.html)\n- R3 Adversarial Network for Cross Model Face Recognition [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FChen_R3_Adversarial_Network_for_Cross_Model_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\")  \n- **MLT**: Face Recognition: A Novel Multi-Level Taxonomy based Survey [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1901.00713 \"arXiv2019\")\n- **CDP**: Consensus-Driven Propagation in Massive Unlabeled Data for Face Recognition [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.01407 \"ECCV2018\") [[code]](https:\u002F\u002Fgithub.com\u002FXiaohangZhan\u002Fface_recognition_framework \"PyTorch\")  [[project]](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FCDP\u002F)\n- **GhostVLAD**: GhostVLAD for set-based face recognition [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.09951 \"ACCV2018\")\n- **DocFace+**: ID Document to Selfie Matching [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.05620 \"arXiv2018\") [[code]](https:\u002F\u002Fgithub.com\u002FseasonSH\u002FDocFace \"TensorFlow\")\n- **2018Survey**: Face Recognition: From Traditional to Deep Learning Methods [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.00116 \"arXiv2018\")\n- **2018Survey**: Deep Facial Expression Recognition: A Survey [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.08348 \"arXiv2018\")\n- **2018Survey**: Deep Face Recognition: A Survey [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.06655 \"arXiv2018\")\n- **SphereFace+(MHE)**: Learning towards Minimum Hyperspherical Energy [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1805.09298 \"arXiv2018\") [[code]](https:\u002F\u002Fgithub.com\u002Fwy1iu\u002Fsphereface-plus \"Caffe\u002FMatlab\")\n- **MobileFace**: A face recognition solution on mobile device [[code]](https:\u002F\u002Fgithub.com\u002FbecauseofAI\u002FMobileFace)\n- **MobileFaceNets**: Efficient CNNs for Accurate Real-Time Face Verification on Mobile Devices [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.07573 \"arXiv2018\") [[code1]](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface \"MXNet\") [[code2]](https:\u002F\u002Fgithub.com\u002FKaleidoZhouYN\u002Fmobilefacenet-caffe \"Caffe\") [[code3]](https:\u002F\u002Fgithub.com\u002Fxsr-ai\u002FMobileFaceNet_TF \"TensorFlow\") [[code4]](https:\u002F\u002Fgithub.com\u002FGRAYKEY\u002Fmobilefacenet_ncnn \"NCNN\")\n- **FaceID**: An implementation of iPhone X's FaceID using face embeddings and siamese networks on RGBD images. [[code]](https:\u002F\u002Fgithub.com\u002Fnormandipalo\u002FfaceID_beta \"Keras\") [[blog]](https:\u002F\u002Ftowardsdatascience.com\u002Fhow-i-implemented-iphone-xs-faceid-using-deep-learning-in-python-d5dbaa128e1d \"Medium\") \n- **InsightFace(ArcFace)**: 2D and 3D Face Analysis Project [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.07698 \"ArcFace: Additive Angular Margin Loss for Deep Face Recognition(arXiv)\") [[code1]](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface \"MXNet\")[![GitHub stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepinsight\u002Finsightface.svg?logo=github&label=Stars)](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface) [[code2]](https:\u002F\u002Fgithub.com\u002Fauroua\u002FInsightFace_TF \"TensorFlow\")\n- **AAM-Softmax(CCL)**: Face Recognition via Centralized Coordinate Learning [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.05678 \"arXiv2018\")\n- **AM-Softmax**: Additive Margin Softmax for Face Verification [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.05599 \"arXiv2018\") [[code1]](https:\u002F\u002Fgithub.com\u002Fhappynear\u002FAMSoftmax \"Caffe\") [[code2]](https:\u002F\u002Fgithub.com\u002FJoker316701882\u002FAdditive-Margin-Softmax \"TensorFlow\")\n- **CosFace**: Large Margin Cosine Loss for Deep Face Recognition [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.09414 \"CVPR2018\") [[code1]](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface \"MXNet\") [[code2]](https:\u002F\u002Fgithub.com\u002Fyule-li\u002FCosFace \"TensorFlow\")\n- **FeatureIncay**: Feature Incay for Representation Regularization [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1705.10284 \"ICLR2018\")\n- **CocoLoss**: Rethinking Feature Discrimination and Polymerization for Large-scale Recognition [[paper]](http:\u002F\u002Fcn.arxiv.org\u002Fabs\u002F1710.00870 \"NIPS2017\") [[code]](https:\u002F\u002Fgithub.com\u002Fsciencefans\u002Fcoco_loss \"Caffe\")\n- **NormFace**: L2 hypersphere embedding for face Verification [[paper]](http:\u002F\u002Fwww.cs.jhu.edu\u002F~alanlab\u002FPubs17\u002Fwang2017normface.pdf \"ACM2017 Multimedia Conference\") [[code]](https:\u002F\u002Fgithub.com\u002Fhappynear\u002FNormFace \"Caffe\")\n- **SphereFace(A-Softmax)**: Deep Hypersphere Embedding for Face Recognition [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2017\u002Fpapers\u002FLiu_SphereFace_Deep_Hypersphere_CVPR_2017_paper.pdf \"CVPR2017\") [[code]](https:\u002F\u002Fgithub.com\u002Fwy1iu\u002Fsphereface \"Caffe\")\n- **L-Softmax**: Large-Margin Softmax Loss for Convolutional Neural Networks [[paper]](http:\u002F\u002Fproceedings.mlr.press\u002Fv48\u002Fliud16.pdf \"ICML2016\") [[code1]](https:\u002F\u002Fgithub.com\u002Fwy1iu\u002FLargeMargin_Softmax_Loss \"Caffe\") [[code2]](https:\u002F\u002Fgithub.com\u002Fluoyetx\u002Fmx-lsoftmax \"MXNet\") [[code3]](https:\u002F\u002Fgithub.com\u002FHiKapok\u002Ftf.extra_losses \"TensorFlow\") [[code4]](https:\u002F\u002Fgithub.com\u002Fauroua\u002FL_Softmax_TensorFlow \"TensorFlow\") [[code5]](https:\u002F\u002Fgithub.com\u002Ftpys\u002Fface-recognition-caffe2 \"Caffe2\") [[code6]](https:\u002F\u002Fgithub.com\u002Famirhfarzaneh\u002Flsoftmax-pytorch \"PyTorch\") [[code7]](https:\u002F\u002Fgithub.com\u002Fjihunchoi\u002Flsoftmax-pytorch \"PyTorch\")\n- **CenterLoss**: A Discriminative Feature Learning Approach for Deep Face Recognition [[paper]](https:\u002F\u002Fydwen.github.io\u002Fpapers\u002FWenECCV16.pdf \"ECCV2016\") [[code1]](https:\u002F\u002Fgithub.com\u002Fydwen\u002Fcaffe-face \"Caffe\") [[code2]](https:\u002F\u002Fgithub.com\u002Fpangyupo\u002Fmxnet_center_loss \"MXNet\") [[code3]](https:\u002F\u002Fgithub.com\u002FShownX\u002Fmxnet-center-loss \"MXNet-Gluon\") [[code4]](https:\u002F\u002Fgithub.com\u002FEncodeTS\u002FTensorFlow_Center_Loss \"TensorFlow\")\n- **OpenFace**: A general-purpose face recognition library with mobile applications [[report]](http:\u002F\u002Felijah.cs.cmu.edu\u002FDOCS\u002FCMU-CS-16-118.pdf \"CMU2016\") [[project]](http:\u002F\u002Fcmusatyalab.github.io\u002Fopenface\u002F) [[code1]](https:\u002F\u002Fgithub.com\u002Fcmusatyalab\u002Fopenface \"Torch\") [[code2]](https:\u002F\u002Fgithub.com\u002Fthnkim\u002FOpenFacePytorch \"PyTorch\")\n- **FaceNet**: A Unified Embedding for Face Recognition and Clustering [[paper]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fpapers\u002FSchroff_FaceNet_A_Unified_2015_CVPR_paper.pdf \"CVPR2015\") [[code]](https:\u002F\u002Fgithub.com\u002Fdavidsandberg\u002Ffacenet \"TensorFlow\")\n- **DeepID3**: DeepID3: Face Recognition with Very Deep Neural Networks [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1502.00873 \"arXiv2015\") \n- **DeepID2+**: Deeply learned face representations are sparse, selective, and robust [[paper]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fpapers\u002FSun_Deeply_Learned_Face_2015_CVPR_paper.pdf \"CVPR2015\")\n- **DeepID2**: Deep Learning Face Representation by Joint Identification-Verification [[paper]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5416-deep-learning-face-representation-by-joint-identification-verification.pdf \"NIPS2014\")\n- **DeepID**: Deep Learning Face Representation from Predicting 10,000 Classes [[paper]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fpapers\u002FSun_Deep_Learning_Face_2014_CVPR_paper.pdf \"CVPR2014\")\n- **DeepFace**: Closing the gap to human-level performance in face verification [[paper]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fpapers\u002FTaigman_DeepFace_Closing_the_2014_CVPR_paper.pdf \"CVPR2014\")\n- **LBP+Joint Bayes**: Bayesian Face Revisited: A Joint Formulation [[paper]](https:\u002F\u002Fs3.amazonaws.com\u002Facademia.edu.documents\u002F31414608\u002FJointBayesian.pdf?AWSAccessKeyId=AKIAIWOWYYGZ2Y53UL3A&Expires=1543656042&Signature=k6LefuQnIC2x8gep7yQTxqKgzus%3D&response-content-disposition=inline%3B%20filename%3DBayesian_Face_Revisited_A_Joint_Formulat.pdf \"ECCV2012\") [[code1]](https:\u002F\u002Fgithub.com\u002Fcyh24\u002FJoint-Bayesian \"Python\") [[code2]](https:\u002F\u002Fgithub.com\u002FMaoXu\u002FJoint_Bayesian \"Matlab\") [[code3]](https:\u002F\u002Fgithub.com\u002FGlasssix\u002Fjoint_bayesian \"C++\u002FC#\")\n- **LBPFace**: Face recognition with local binary patterns [[paper]](https:\u002F\u002Fpdfs.semanticscholar.org\u002F3242\u002F0c65f8ef0c5bd83b14c8ae662cbce73e6781.pdf \"ECCV2004\") [[code]](https:\u002F\u002Fdocs.opencv.org\u002F2.4\u002Fmodules\u002Fcontrib\u002Fdoc\u002Ffacerec\u002Ffacerec_tutorial.html \"OpenCV\")\n- **FisherFace(LDA)**: Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection [[paper]](https:\u002F\u002Fapps.dtic.mil\u002Fdtic\u002Ftr\u002Ffulltext\u002Fu2\u002F1015508.pdf \"TPAMI1997\") [[code]](https:\u002F\u002Fdocs.opencv.org\u002F2.4\u002Fmodules\u002Fcontrib\u002Fdoc\u002Ffacerec\u002Ffacerec_tutorial.html \"OpenCV\")\n- **EigenFace(PCA)**: Face recognition using eigenfaces [[paper]](http:\u002F\u002Fwww.cs.ucsb.edu\u002F~mturk\u002FPapers\u002Fmturk-CVPR91.pdf \"CVPR1991\") [[code]](https:\u002F\u002Fdocs.opencv.org\u002F2.4\u002Fmodules\u002Fcontrib\u002Fdoc\u002Ffacerec\u002Ffacerec_tutorial.html \"OpenCV\")\n\n## 🔖 Face Detection\n- **RetinaFace**: Single-stage Dense Face Localisation in the Wild [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.00641 \"arXiv2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface\u002Ftree\u002Fmaster\u002FRetinaFace \"MXNet\")\n- Group Sampling for Scale Invariant Face Detection [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FMing_Group_Sampling_for_Scale_Invariant_Face_Detection_CVPR_2019_paper.pdf \"CVPR2019\")\n- **FA-RPN**: Floating Region Proposals for Face Detection [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FNajibi_FA-RPN_Floating_Region_Proposals_for_Face_Detection_CVPR_2019_paper.pdf \"CVPR2019\")\n- **SFA**: Small Faces Attention Face Detector [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.08402 \"SPIC2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fshiluo1990\u002FSFA \"Caffe\")\n- **ISRN**: Improved Selective Refinement Network for Face Detection [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1901.06651 \"arXiv2019\")\n- **DSFD**: Dual Shot Face Detector [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.10220 \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002FTencentYoutuResearch\u002FFaceDetection-DSFD \"PyTorch\")\n- **PyramidBox++**: High Performance Detector for Finding Tiny Face [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.00386 \"arXiv2019\")\n- **VIM-FD**: Robust and High Performance Face Detector [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1901.02350 \"arXiv2019\")\n- **SHF**: Robust Face Detection via Learning Small Faces on Hard Images [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.11662 \"arXiv2018\") [[code]](https:\u002F\u002Fgithub.com\u002Fbairdzhang\u002Fsmallhardface \"Caffe\")\n- **SRN**: Selective Refinement Network for High Performance Face Detection [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.02693 \"AAAI2019\")\n- **SFDet**: Single-Shot Scale-Aware Network for Real-Time Face Detection [[paper]](https:\u002F\u002Flink.springer.com\u002Fepdf\u002F10.1007\u002Fs11263-019-01159-3?author_access_token=Jjgl-u1CAXPmSKWDljfSBfe4RwlQNchNByi7wbcMAY7Vwo_nrkuFMElF6YSQ0We34tUs42D0dyurcBAD0sJP66n6GBanVgA9qsuvh4Y_Bjf3E_n9_croQ4esS882srfHyUz-L96pU3gu_M30Kk6_XQ%3D%3D \"IJCV2019\")\n- **HyperFace**: A Deep Multi-task Learning Framework for Face Detection, Landmark Localization, Pose Estimation, and Gender Recognition [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1603.01249 \"TPAMI2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fmaharshi95\u002FHyperFace \"TensorFlow\")\n- **PyramidBox**: A Context-assisted Single Shot Face Detector [[paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1803.07737.pdf \"arXiv2018\") [[code]](https:\u002F\u002Fgithub.com\u002FPaddlePaddle\u002Fmodels\u002Ftree\u002F2a6b7dc92f04815f0b298e59030cb779dd0e038c\u002Ffluid\u002Fface_detction \"PaddlePaddle\")\n- **PCN**: Real-Time Rotation-Invariant Face Detection with Progressive Calibration Networks [[paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.06039.pdf \"CVPR2018\") [[code]](https:\u002F\u002Fgithub.com\u002FJack-CV\u002FPCN \"C++\") \n- **S³FD**: Single Shot Scale-invariant Face Detector [[paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1708.05237.pdf \"arXiv2017\") [[code]](https:\u002F\u002Fgithub.com\u002Fsfzhang15\u002FSFD \"Caffe\")\n- **SSH**: Single Stage Headless Face Detector [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2017\u002Fpapers\u002FNajibi_SSH_Single_Stage_ICCV_2017_paper.pdf \"ICCV2017\") [[code]](https:\u002F\u002Fgithub.com\u002Fmahyarnajibi\u002FSSH \"Caffe\")\n- **FaceBoxes**: A CPU Real-time Face Detector with High Accuracy [[paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1708.05234.pdf \"IJCB2017\")[[code1]](https:\u002F\u002Fgithub.com\u002Fzeusees\u002FFaceBoxes \"Caffe\") [[code2]](https:\u002F\u002Fgithub.com\u002Flxg2015\u002Ffaceboxes \"PyTorch\")\n- **TinyFace**: Finding Tiny Faces [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2017\u002Fpapers\u002FHu_Finding_Tiny_Faces_CVPR_2017_paper.pdf \"CVPR2017\") [[project]](https:\u002F\u002Fwww.cs.cmu.edu\u002F~peiyunh\u002Ftiny\u002F) [[code1]](https:\u002F\u002Fgithub.com\u002Fpeiyunh\u002Ftiny \"MatConvNet\") [[code2]](https:\u002F\u002Fgithub.com\u002Fchinakook\u002Fhr101_mxnet \"MXNet\") [[code3]](https:\u002F\u002Fgithub.com\u002Fcydonia999\u002FTiny_Faces_in_Tensorflow \"TensorFlow\")\n- **MTCNN**: Joint Face Detection and Alignment using Multi-task Cascaded Convolutional Networks [[paper]](https:\u002F\u002Fkpzhang93.github.io\u002FMTCNN_face_detection_alignment\u002Fpaper\u002Fspl.pdf \"SPL2016\") [[project]](https:\u002F\u002Fkpzhang93.github.io\u002FMTCNN_face_detection_alignment\u002F) [[code1]](https:\u002F\u002Fgithub.com\u002Fkpzhang93\u002FMTCNN_face_detection_alignment \"Caffe\") [[code2]](https:\u002F\u002Fgithub.com\u002FCongWeilin\u002Fmtcnn-caffe \"Caffe\") [[code3]](https:\u002F\u002Fgithub.com\u002FforeverYoungGitHub\u002FMTCNN \"Caffe\") [[code4]](https:\u002F\u002Fgithub.com\u002FSeanlinx\u002Fmtcnn \"MXNet\") [[code5]](https:\u002F\u002Fgithub.com\u002Fpangyupo\u002Fmxnet_mtcnn_face_detection \"MXNet\") [[code6]](https:\u002F\u002Fgithub.com\u002FTropComplique\u002Fmtcnn-pytorch \"PyTorch\") [[code7]](https:\u002F\u002Fgithub.com\u002FAITTSMD\u002FMTCNN-Tensorflow \"TensorFlow\")\n- **NPD**: A Fast and Accurate Unconstrained Face Detector [[paper]](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fusers\u002Fscliao\u002Fpapers\u002FLiao-PAMI15-NPD.pdf \"TPAMI2015\") [[code]](https:\u002F\u002Fgithub.com\u002Fwincle\u002FNPD \"C++\") [[project]](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fusers\u002Fscliao\u002Fprojects\u002Fnpdface\u002Findex.html)\n- **PICO**: Object Detection with Pixel Intensity Comparisons Organized in Decision Trees [[paper]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1305.4537.pdf \"arXiv2014\") [[code]](https:\u002F\u002Fgithub.com\u002Fnenadmarkus\u002Fpico \"C\")\n- **libfacedetection**: A fast binary library for face detection and face landmark detection in images. [[code]](https:\u002F\u002Fgithub.com\u002FShiqiYu\u002Flibfacedetection \"C++\")\n- **SeetaFaceEngine**: SeetaFace Detection, SeetaFace Alignment and SeetaFace Identification [[code]](https:\u002F\u002Fgithub.com\u002Fseetaface\u002FSeetaFaceEngine \"C++\")\n\n## 🔖 Face Landmark\n- **DeCaFA**: Deep Convolutional Cascade for Face Alignment in the Wild [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.02549)\n- **AWing**: Adaptive Wing Loss for Robust Face Alignment via Heatmap Regression [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.07399 \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fprotossw512\u002FAdaptiveWingLoss \"PyTorch\")\n- **KDN**: Face Alignment With Kernel Density Deep Neural Network [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FChen_Face_Alignment_With_Kernel_Density_Deep_Neural_Network_ICCV_2019_paper.pdf \"ICCV2019\")\n- **Semantic Alignment**: Finding Semantically Consistent Ground-Truth for Facial Landmark Detection [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FLiu_Semantic_Alignment_Finding_Semantically_Consistent_Ground-Truth_for_Facial_Landmark_Detection_CVPR_2019_paper.pdf \"CVPR2019\")\n- Robust Facial Landmark Detection via Occlusion-Adaptive Deep Networks [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhu_Robust_Facial_Landmark_Detection_via_Occlusion-Adaptive_Deep_Networks_CVPR_2019_paper.pdf \"CVPR2019\")\n- **PFLD**: A Practical Facial Landmark Detector [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.10859 \"arXiv2019\") [[project]](https:\u002F\u002Fsites.google.com\u002Fview\u002Fxjguo\u002Ffld) [[code]](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1n1uZPbM9Wz052aVnlc_3L4gjQHiwfj4B\u002Fview \"APK\")\n- **PRNet**: Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Network [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ECCV_2018\u002Fpapers\u002FYao_Feng_Joint_3D_Face_ECCV_2018_paper.pdf \"ECCV2018\") [[code]](https:\u002F\u002Fgithub.com\u002FYadiraF\u002FPRNet \"TensorFlow\")\n- **LAB**: Look at Boundary: A Boundary-Aware Face Alignment Algorithm [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FWu_Look_at_Boundary_CVPR_2018_paper.pdf \"CVPR2018\") [[project]](https:\u002F\u002Fwywu.github.io\u002Fprojects\u002FLAB\u002FLAB.html) [[code]](https:\u002F\u002Fgithub.com\u002Fwywu\u002FLAB \"Caffe\")\n- **Face-Alignment**: How far are we from solving the 2D & 3D Face Alignment problem? (and a dataset of 230,000 3D facial landmarks)  [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2017\u002Fpapers\u002FBulat_How_Far_Are_ICCV_2017_paper.pdf \"ICCV2017\") [[project]](https:\u002F\u002Fadrianbulat.com\u002Fface-alignment) [[code1]](https:\u002F\u002Fgithub.com\u002F1adrianb\u002Fface-alignment \"PyTorch\") [[code2]](https:\u002F\u002Fgithub.com\u002F1adrianb\u002F2D-and-3D-face-alignment \"Torch7\")\n- **ERT**: One Millisecond Face Alignment with an Ensemble of Regression Trees [[paper]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fpapers\u002FKazemi_One_Millisecond_Face_2014_CVPR_paper.pdf \"CVPR2014\") [[code]](http:\u002F\u002Fdlib.net\u002Fimaging.html \"Dlib\")\n\n## 🔖 Face Clustering\n- **BCL**: Video Face Clustering With Unknown Number of Clusters [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.03381 \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fmakarandtapaswi\u002FBallClustering_ICCV2019 \"PyTorch\")\n- **LinkageFace**: Linkage Based Face Clustering via Graph Convolution Network [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.11306 \"CVPR2019\")\n- **LTC**: Learning to Cluster Faces on an Affinity Graph [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FYang_Learning_to_Cluster_Faces_on_an_Affinity_Graph_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fyl-1993\u002Flearn-to-cluster \"PyTorch\") \n\n## 🔖 Face Expression \n- **FECNet**: A Compact Embedding for Facial Expression Similarity [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FVemulapalli_A_Compact_Embedding_for_Facial_Expression_Similarity_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002FGerardLiu96\u002FFECNet \"Keras\")\n- **LBVCNN**: Local Binary Volume Convolutional Neural Network for Facial Expression Recognition from Image Sequences [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.07647 \"arXiv2019\")\n\n## 🔖 Face Action\n- Joint Representation and Estimator Learning for Facial Action Unit Intensity Estimation [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhang_Joint_Representation_and_Estimator_Learning_for_Facial_Action_Unit_Intensity_CVPR_2019_paper.pdf \"CVPR2019\")\n- Local Relationship Learning With Person-Specific Shape Regularization for Facial Action Unit Detection [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FNiu_Local_Relationship_Learning_With_Person-Specific_Shape_Regularization_for_Facial_Action_CVPR_2019_paper.pdf \"CVPR2019\")\n- **TCAE**: Self-Supervised Representation Learning From Videos for Facial Action Unit Detection [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FLi_Self-Supervised_Representation_Learning_From_Videos_for_Facial_Action_Unit_Detection_CVPR_2019_paper.pdf \"CVPR2019 Oral\") [[code]](https:\u002F\u002Fgithub.com\u002Fmysee1989\u002FTCAE \"PyTorch\")\n- **JAANet**: Deep Adaptive Attention for Joint Facial Action Unit Detection and Face Alignment [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ECCV_2018\u002Fpapers\u002FZhiwen_Shao_Deep_Adaptive_Attention_ECCV_2018_paper.pdf \"ECCV2018\") [[code]](https:\u002F\u002Fgithub.com\u002FZhiwenShao\u002FJAANet \"Caffe\")\n\n## 🔖 Face 3D\n- **DF2Net**: A Dense-Fine-Finer Network for Detailed 3D Face Reconstruction [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FZeng_DF2Net_A_Dense-Fine-Finer_Network_for_Detailed_3D_Face_Reconstruction_ICCV_2019_paper.pdf \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fxiaoxingzeng\u002FDF2Net \"PyTorch\")\n- Semi-Supervised Monocular 3D Face Reconstruction With End-to-End Shape-Preserved Domain Transfer [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FPiao_Semi-Supervised_Monocular_3D_Face_Reconstruction_With_End-to-End_Shape-Preserved_Domain_Transfer_ICCV_2019_paper.pdf \"ICCV2019\")\n- **3DFC**: 3D Face Modeling From Diverse Raw Scan Data [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.04943 \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fliuf1990\u002F3DFC \"PyTorch\")\n- **2DASL**: Joint 3D Face Reconstruction and Dense Face Alignment from A Single Image with 2D-Assisted Self-Supervised Learning [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.09359 \"arXiv2019\") [[code]](https:\u002F\u002Fgithub.com\u002FXgTu\u002F2DASL \"PyTorch & Matlab\")\n- **MVF-Net**: Multi-View 3D Face Morphable Model Regression [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FWu_MVF-Net_Multi-View_3D_Face_Morphable_Model_Regression_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002FFanziapril\u002Fmvfnet \"PyTorch\")\n- Dense 3D Face Decoding Over 2500FPS: Joint Texture & Shape Convolutional Mesh Decoders [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhou_Dense_3D_Face_Decoding_Over_2500FPS_Joint_Texture__Shape_CVPR_2019_paper.pdf \"CVPR2019\")\n- Towards High-Fidelity Nonlinear 3D Face Morphable Model [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FTran_Towards_High-Fidelity_Nonlinear_3D_Face_Morphable_Model_CVPR_2019_paper.pdf \"CVPR2019\") [[project]](http:\u002F\u002Fcvlab.cse.msu.edu\u002Fproject-nonlinear-3dmm.html)\n- Combining 3D Morphable Models: A Large Scale Face-And-Head Model [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FPloumpis_Combining_3D_Morphable_Models_A_Large_Scale_Face-And-Head_Model_CVPR_2019_paper.pdf \"CVPR2019\") \n- Disentangled Representation Learning for 3D Face Shape [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FJiang_Disentangled_Representation_Learning_for_3D_Face_Shape_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002FzihangJiang\u002FDR-Learning-for-3D-Face \"Keras\")\n- Self-Supervised Adaptation of High-Fidelity Face Models for Monocular Performance Tracking [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FYoon_Self-Supervised_Adaptation_of_High-Fidelity_Face_Models_for_Monocular_Performance_Tracking_CVPR_2019_paper.pdf \"CVPR2019\")\n- **MMFace**: A Multi-Metric Regression Network for Unconstrained Face Reconstruction [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FYi_MMFace_A_Multi-Metric_Regression_Network_for_Unconstrained_Face_Reconstruction_CVPR_2019_paper.pdf \"CVPR2019\")\n- **RingNet**: Learning to Regress 3D Face Shape and Expression From an Image Without 3D Supervision [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FSanyal_Learning_to_Regress_3D_Face_Shape_and_Expression_From_an_CVPR_2019_paper.pdf \"CVPR2019\")  [[code]](https:\u002F\u002Fgithub.com\u002Fsoubhiksanyal\u002FRingNet \"TensorFlow\") [[project]](https:\u002F\u002Fringnet.is.tue.mpg.de\u002F)\n- Boosting Local Shape Matching for Dense 3D Face Correspondence [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FFan_Boosting_Local_Shape_Matching_for_Dense_3D_Face_Correspondence_CVPR_2019_paper.pdf \"CVPR2019\")\n- **FML**: Face Model Learning From Videos [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FTewari_FML_Face_Model_Learning_From_Videos_CVPR_2019_paper.pdf \"CVPR2019\")\n\n## 🔖 Face GAN\n#### Face Character\n- Face-to-Parameter Translation for Game Character Auto-Creation [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.01064 \"ICCV2019\")\n---\n#### Face Editing\n- **SC-FEGAN**: Face Editing Generative Adversarial Network With User's Sketch and Color [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.06838 \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002Frun-youngjoo\u002FSC-FEGAN \"TensorFlow\")\n---\n#### Face De-Occlusion\n- Face De-Occlusion Using 3D Morphable Model and Generative Adversarial Network [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.06109 \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fxweiyuan\u002FFace-de-occlusion-using-3D-morphable-model-and-generative-adversarial-network)\n---\n#### Face Aging  \n- Automatic Face Aging in Videos via Deep Reinforcement Learning [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FDuong_Automatic_Face_Aging_in_Videos_via_Deep_Reinforcement_Learning_CVPR_2019_paper.pdf \"CVPR2019\") [[blog]](https:\u002F\u002Fwww.fastcompany.com\u002F90314606\u002Fthis-new-ai-tool-makes-creepily-realistic-videos-of-faces-in-the-future)\n- Attribute-Aware Face Aging With Wavelet-Based Generative Adversarial Networks [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FLiu_Attribute-Aware_Face_Aging_With_Wavelet-Based_Generative_Adversarial_Networks_CVPR_2019_paper.pdf \"CVPR2019\")\n- **SAGAN**:Generative Adversarial Network with Spatial Attention for Face Attribute Editing [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ECCV_2018\u002Fpapers\u002FGang_Zhang_Generative_Adversarial_Network_ECCV_2018_paper.pdf \"ECCV2018\") [[code]](https:\u002F\u002Fgithub.com\u002Felvisyjlin\u002FSpatialAttentionGAN \"PyTorch\")\n---\n#### Face Drawing \n- **APDrawingGAN**: Generating Artistic Portrait Drawings From Face Photos With Hierarchical GANs [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FYi_APDrawingGAN_Generating_Artistic_Portrait_Drawings_From_Face_Photos_With_Hierarchical_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fyiranran\u002FAPDrawingGAN \"PyTorch\")\n---\n#### Face Generation\n- **StyleGAN**: A Style-Based Generator Architecture for Generative Adversarial Networks [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FKarras_A_Style-Based_Generator_Architecture_for_Generative_Adversarial_Networks_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan \"TensorFlow\") [[dataset]](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset \"FFHQ\")\n---\n#### Face Makeup\n- **BeautyGAN**: Instance-level Facial Makeup Transfer with Deep Generative Adversarial Network [[paper]](http:\u002F\u002Fliusi-group.com\u002Fpdf\u002FBeautyGAN-camera-ready_2.pdf \"Multimedia Conference, ACM2018\") [[code]](https:\u002F\u002Fgithub.com\u002FHonlan\u002FBeautyGAN \"TensorFlow\") [[project]](http:\u002F\u002Fliusi-group.com\u002Fprojects\u002FBeautyGAN) [[poster]](http:\u002F\u002Fliusi-group.com\u002Fpdf\u002FBeautyGAN-camera-ready_2_poster.pdf)\n---\n#### Face Swap\n- **FSGAN**: Subject Agnostic Face Swapping and Reenactment [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.05932 \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002FYuvalNirkin) [[project]](https:\u002F\u002Fnirkin.com\u002Ffsgan\u002F)\n- **Faceswap**: A tool that utilizes deep learning to recognize and swap faces in pictures and videos [[code1]](https:\u002F\u002Fgithub.com\u002Fdeepfakes\u002Ffaceswap \"TensorFlow\") [[code2]](https:\u002F\u002Fgithub.com\u002Fiperov\u002FDeepFaceLab \"TensorFlow\u002FKeras\")\n- **FUNIT**: Few-Shot Unsupervised Image-to-Image Translation  [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.01723 \"arXiv2019\") [[code]](https:\u002F\u002Fgithub.com\u002FNVlabs\u002FFUNIT \"PyTorch\") [[project]](https:\u002F\u002Fnvlabs.github.io\u002FFUNIT\u002F)\n---\n#### Face Other\n- Unsupervised Face Normalization With Extreme Pose and Expression in the Wild [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FQian_Unsupervised_Face_Normalization_With_Extreme_Pose_and_Expression_in_the_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fmx54039q\u002Ffnm \"TensorFlow\")\n- **GANFIT**: Generative Adversarial Network Fitting for High Fidelity 3D Face Reconstruction [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FGecer_GANFIT_Generative_Adversarial_Network_Fitting_for_High_Fidelity_3D_Face_CVPR_2019_paper.pdf \"CVPR2019\") [[project]](https:\u002F\u002Fgithub.com\u002Fbarisgecer\u002FGANFit)\n- **HF-PIM**: Learning a High Fidelity Pose Invariant Model for High-resolution Face Frontalization [[paper]](http:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F7551-learning-a-high-fidelity-pose-invariant-model-for-high-resolution-face-frontalization.pdf \"NIPS2018\")\n- **Super-FAN**: Integrated facial landmark localization and super-resolution of real-world low resolution faces in arbitrary poses with GANs [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FBulat_Super-FAN_Integrated_Facial_CVPR_2018_paper.pdf \"CVPR2018 Spotlight\")\n- **GANimation**: Anatomically-aware Facial Animation from a Single Image [[paper]](https:\u002F\u002Fwww.albertpumarola.com\u002Fpublications\u002Ffiles\u002Fpumarola2018ganimation.pdf \"ECCV2018 Oral,Best Paper Award Honorable Mention\") [[project]](https:\u002F\u002Fwww.albertpumarola.com\u002Fresearch\u002FGANimation\u002Findex.html) [[code]](https:\u002F\u002Fgithub.com\u002Falbertpumarola\u002FGANimation \"PyTorch\")\n- **StarGAN**: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FChoi_StarGAN_Unified_Generative_CVPR_2018_paper.pdf \"CVPR2018\")\n[[code]](https:\u002F\u002Fgithub.com\u002Fyunjey\u002FStarGAN \"PyTorch\")\n- **PGAN**: Progressive Growing of GANs for Improved Quality, Stability, and Variation [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1710.10196 \"ICLR2018\")\n[[code1]](https:\u002F\u002Fgithub.com\u002Ftkarras\u002Fprogressive_growing_of_gans \"TensorFlow\") [[code2]](https:\u002F\u002Fgithub.com\u002Fgithub-pengge\u002FPyTorch-progressive_growing_of_gans \"PyTorch\")\n\n## 🔖 Face Deblurring\n- Face Video Deblurring Using 3D Facial Priors [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FRen_Face_Video_Deblurring_Using_3D_Facial_Priors_ICCV_2019_paper.pdf \"ICCV2019\") [[code]](https:\u002F\u002Fgithub.com\u002Frwenqi\u002F3Dfacedeblurring \"TensorFlow\")\n\n## 🔖 Face Super-Resolution\n- **PFSR**: Progressive Face Super-Resolution via Attention to Facial Landmark [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.08239 \"BMVC2019\") [[code]](https:\u002F\u002Fgithub.com\u002FDeokyunKim\u002FProgressive-Face-Super-Resolution \"PyTorch\")\n\n## 🔖 Face Manipulation\n- **Make a Face**: Towards Arbitrary High Fidelity Face Manipulation [[paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.07191 \"ICCV2019\")\n- 3D Guided Fine-Grained Face Manipulation [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FGeng_3D_Guided_Fine-Grained_Face_Manipulation_CVPR_2019_paper.pdf \"CVPR2019\")\n- **SemanticComponent**: Semantic Component Decomposition for Face Attribute Manipulation [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FChen_Semantic_Component_Decomposition_for_Face_Attribute_Manipulation_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fyingcong\u002FSemanticComponent) [[demo]](http:\u002F\u002Fappsrv.cse.cuhk.edu.hk\u002F~ycchen\u002Fdemos\u002Fsemantic_component.mp4)\n\n## 🔖 Face Anti-Spoofing\n- **Dataset and Benchmark**: A Dataset and Benchmark for Large-Scale Multi-Modal Face Anti-Spoofing [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhang_A_Dataset_and_Benchmark_for_Large-Scale_Multi-Modal_Face_Anti-Spoofing_CVPR_2019_paper.pdf \"CVPR2019\") [[poster]](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fusers\u002Fsfzhang\u002FShifeng%20Zhang's%20Homepage_files\u002FCVPR2019_CASIA-SURF_Poster.pdf) [[dataset]](https:\u002F\u002Fsites.google.com\u002Fqq.com\u002Fchalearnfacespoofingattackdete\u002F)\n- Deep Tree Learning for Zero-Shot Face Anti-Spoofing [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FLiu_Deep_Tree_Learning_for_Zero-Shot_Face_Anti-Spoofing_CVPR_2019_paper.pdf \"CVPR2019 Oral\") \n\n## 🔖Face Adversarial Attack\n- Decorrelated Adversarial Learning for Age-Invariant Face Recognition [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FWang_Decorrelated_Adversarial_Learning_for_Age-Invariant_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\") \n- Multi-Adversarial Discriminative Deep Domain Generalization for Face Presentation Attack Detection [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FShao_Multi-Adversarial_Discriminative_Deep_Domain_Generalization_for_Face_Presentation_Attack_Detection_CVPR_2019_paper.pdf \"CVPR2019\") \n- Efficient Decision-Based Black-Box Adversarial Attacks on Face Recognition [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FDong_Efficient_Decision-Based_Black-Box_Adversarial_Attacks_on_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\") \n\n## 🔖 Face Cross-Modal\n- **From Inference to Generation**: End-to-end Fully Self-supervised Generation of Human Face from Speech [[paper]](https:\u002F\u002Fopenreview.net\u002Fpdf?id=H1guaREYPr \"ICLR2020\")\n- **FRV**: Face Reconstruction from Voice using Generative Adversarial Networks [[paper]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F8768-face-reconstruction-from-voice-using-generative-adversarial-networks.pdf \"NeurIPS2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fcmu-mlsp\u002Freconstructing_faces_from_voices \"PyTorch\") [[poster]](https:\u002F\u002Fydwen.github.io\u002Fpapers\u002FWenNeurIPS19-poster.pdf)\n- **Speech2Face**: Learning the Face Behind a Voice [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FOh_Speech2Face_Learning_the_Face_Behind_a_Voice_CVPR_2019_paper.pdf \"CVPR2019\") [[project]](https:\u002F\u002Fspeech2face.github.io\u002F)\n- **JFDFMR**: Joint Face Detection and Facial Motion Retargeting for Multiple Faces [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FChaudhuri_Joint_Face_Detection_and_Facial_Motion_Retargeting_for_Multiple_Faces_CVPR_2019_paper.pdf \"CVPR2019\")\n- **ATVGnet**: Hierarchical Cross-Modal Talking Face Generation With Dynamic Pixel-Wise Loss [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FChen_Hierarchical_Cross-Modal_Talking_Face_Generation_With_Dynamic_Pixel-Wise_Loss_CVPR_2019_paper.pdf \"CVPR2019\")  [[code]](https:\u002F\u002Fgithub.com\u002Flelechen63\u002FATVGnet \"PyTorch\")\n\n## 🔖 Face Capture\n- High-Quality Face Capture Using Anatomical Muscles [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FBao_High-Quality_Face_Capture_Using_Anatomical_Muscles_CVPR_2019_paper.pdf \"CVPR2019\")\n- Monocular Total Capture: Posing Face, Body, and Hands in the Wild [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FXiang_Monocular_Total_Capture_Posing_Face_Body_and_Hands_in_the_CVPR_2019_paper.pdf \"CVPR2019\")  [[code]](https:\u002F\u002Fgithub.com\u002FCMU-Perceptual-Computing-Lab\u002FMonocularTotalCapture) [[project]](http:\u002F\u002Fdomedb.perception.cs.cmu.edu\u002Fmtc.html)\n- Expressive Body Capture: 3D Hands, Face, and Body From a Single Image [[paper]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FPavlakos_Expressive_Body_Capture_3D_Hands_Face_and_Body_From_a_CVPR_2019_paper.pdf \"CVPR2019\") [[code]](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplify-x \"PyTorch\") [[project]](https:\u002F\u002Fsmpl-x.is.tue.mpg.de\u002F)\n\n## :hammer: Face Lib and Tool\n- **Dlib** [[url]](http:\u002F\u002Fdlib.net\u002Fimaging.html \"Image Processing\") [[github]](https:\u002F\u002Fgithub.com\u002Fdavisking\u002Fdlib \"master\")\n- **OpenCV** [[docs]](https:\u002F\u002Fdocs.opencv.org \"All Versions\") [[github]](https:\u002F\u002Fgithub.com\u002Fopencv\u002Fopencv\u002F \"master\")\n- **Face3D** [[github]](https:\u002F\u002Fgithub.com\u002FYadiraF\u002Fface3d \"master\")\n\n---\n# \u003Cp align=\"center\">:boom:**Big Bang**:boom:\u003C\u002Fp>\n\n#### \u003Cp align=\"center\">**Receptive Field Is Natural Anchor**\u003C\u002Fp>\n#### \u003Cp align=\"center\">**Receptive Field Is All You Need**\u003C\u002Fp>\n\u003Cp align=\"center\">2K real-time detection is so easy!\u003C\u002Fp>\n\n\u003Cdiv align=\"center\">\u003Cimg width=\"1280\" height=\"auto\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FbecauseofAI_awesome-face_readme_d8b4e880ba80.gif\"\u002F>\u003C\u002Fdiv>  \n\n#### \u003Cp align=\"center\">[[Paper]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.10633) [[MXNet]](https:\u002F\u002Fgithub.com\u002FYonghaoHe\u002FA-Light-and-Fast-Face-Detector-for-Edge-Devices) [[PyTorch]](https:\u002F\u002Fgithub.com\u002FbecauseofAI\u002Flffd-pytorch) [![GitHub stars](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYonghaoHe\u002FA-Light-and-Fast-Face-Detector-for-Edge-Devices.svg?logo=github&label=Stars)](https:\u002F\u002Fgithub.com\u002FYonghaoHe\u002FA-Light-and-Fast-Face-Detector-for-Edge-Devices)\u003C\u002Fp>\n","# HelloFace [![Awesome HelloFace 中提及](https:\u002F\u002Fawesome.re\u002Fmentioned-badge.svg)](https:\u002F\u002Fgithub.com\u002FbecauseofAI\u002FHelloFace)   \n一个优秀的面部技术资源库。（**持续更新中**）  \n\n## :trophy: SOTA\n[![PWC](https:\u002F\u002Fimg.shields.io\u002Fendpoint.svg?url=https:\u002F\u002Fpaperswithcode.com\u002Fbadge\u002Ftinaface-strong-but-simple-baseline-for-face\u002Fface-detection-on-wider-face-hard)](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fface-detection-on-wider-face-hard?p=tinaface-strong-but-simple-baseline-for-face)  \n[![PWC](https:\u002F\u002Fimg.shields.io\u002Fendpoint.svg?url=https:\u002F\u002Fpaperswithcode.com\u002Fbadge\u002Fsubpixel-heatmap-regression-for-facial\u002Fface-alignment-on-wflw)](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fface-alignment-on-wflw?p=subpixel-heatmap-regression-for-facial)  \n[![PWC](https:\u002F\u002Fimg.shields.io\u002Fendpoint.svg?url=https:\u002F\u002Fpaperswithcode.com\u002Fbadge\u002Fdeep-polynomial-neural-networks\u002Fface-verification-on-megaface)](https:\u002F\u002Fpaperswithcode.com\u002Fsota\u002Fface-verification-on-megaface?p=deep-polynomial-neural-networks)\n\n## :dart: 亮点\n\u003Cdiv align=\"center\">\u003Cimg width=\"1280\" height=\"auto\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FbecauseofAI_awesome-face_readme_d0ab636e646b.jpg\"\u002F>\u003C\u002Fdiv>\n\u003Cp align=\"center\">人脸检测\u003C\u002Fp>  \n\n\u003Cdiv align=\"center\">\u003Cimg width=\"1280\" height=\"auto\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FbecauseofAI_awesome-face_readme_fa47efb31ecd.jpg\"\u002F>\u003C\u002Fdiv>\n\u003Cp align=\"center\">人脸对齐\u003C\u002Fp>  \n\n\u003Cdiv align=\"center\">\u003Cimg width=\"1280\" height=\"auto\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FbecauseofAI_awesome-face_readme_37c1e099f77f.jpg\"\u002F>\u003C\u002Fdiv>\n\u003Cp align=\"center\">人脸识别\u003C\u002Fp>  \n\n## :computer: 官网\nhttps:\u002F\u002FbecauseofAI.github.io\u002FHelloFace（欢迎作为贡献者通过 Pull Request 维护该网站。）\n\n## :bookmark_tabs: 内容\n- [近期更新](#recent-update)\n  - [2022-04-05](#2022-04-05)\n  - [2020-10-02](#2020-10-02)\n  - [2020-01-26](#2020-01-26)\n  - [2019-07-11](#2019-07-11)\n  - [2019-04-06](#2019-04-06)\n  - [2019-01-12](#2019-01-12)\n  - [2018-12-01](#2018-12-01)\n  - [2018-07-21](#2018-07-21)\n  - [2018-04-20](#2018-04-20)\n  - [2018-03-28](#2018-03-28)\n- [人脸基准与数据集](#face-benchmark-and-dataset)\n  - [人脸识别数据](#face-recognition-data)\n  - [人脸检测数据](#face-detection-data)\n  - [人脸关键点数据](#face-landmark-data)\n  - [人脸属性数据](#face-attribute-data)\n- [人脸识别](#face-recognition)\n- [人脸检测](#face-detection)\n- [人脸关键点](#face-landmark)\n- [人脸聚类](#face-clustering)\n- [人脸表情](#face-expression)\n- [人脸动作](#face-action)\n- [人脸3D](#face-3d)\n- [人脸GAN](#face-gan)\n  - [人脸角色生成](#face-character)\n  - [人脸编辑](#face-editing)\n  - [人脸去遮挡](#face-de-occlusion)\n  - [人脸年龄变换](#face-aging)\n  - [人脸绘画](#face-drawing)\n  - [人脸生成](#face-generation)\n  - [人脸化妆](#face-makeup)\n  - [人脸交换](#face-swap)\n  - [其他人脸相关](#face-other)\n- [人脸去模糊](#face-deblurring)\n- [人脸超分辨率](#face-super-resolution)\n- [人脸操控](#face-manipulation)\n- [人脸防伪](#face-anti-spoofing)\n- [人脸对抗攻击](#face-adversarial-attack)\n- [跨模态人脸](#face-cross-modal)\n- [人脸采集](#face-capture)\n- [人脸库与工具](#face-lib-and-tool)\n\n## 👋 最新更新\n###### 2022-04-05\n**CVPR2021**  \n- **VirFace**: 通过无标签浅层数据增强人脸识别 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_VirFace_Enhancing_Face_Recognition_via_Unlabeled_Shallow_Data_CVPR_2021_paper.pdf)  \n- **MagFace**: 一种用于人脸识别和质量评估的通用表示 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FMeng_MagFace_A_Universal_Representation_for_Face_Recognition_and_Quality_Assessment_CVPR_2021_paper.pdf)\n- 面向深度人脸识别的变分原型学习 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FDeng_Variational_Prototype_Learning_for_Deep_Face_Recognition_CVPR_2021_paper.pdf)\n- 面向未见域下人脸识别的跨域相似性学习 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FFaraki_Cross-Domain_Similarity_Learning_for_Face_Recognition_in_Unseen_Domains_CVPR_2021_paper.pdf)\n- 虚拟全连接层：在有限计算资源下训练大规模人脸识别数据集 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Virtual_Fully-Connected_Layer_Training_a_Large-Scale_Face_Recognition_Dataset_With_CVPR_2021_paper.pdf)\n- 通过群体自适应分类器缓解人脸识别偏见 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FGong_Mitigating_Face_Recognition_Bias_via_Group_Adaptive_Classifier_CVPR_2021_paper.pdf)\n- 用于人脸识别的极端姿态伪人脸生成 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FWang_Pseudo_Facial_Generation_With_Extreme_Poses_for_Face_Recognition_CVPR_2021_paper.pdf)  \n- 用于野外大规模人脸识别的动态类别队列 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Dynamic_Class_Queue_for_Large_Scale_Face_Recognition_in_the_CVPR_2021_paper.pdf)\n- 利用生成模型提升对抗补丁在人脸识别上的迁移性 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FXiao_Improving_Transferability_of_Adversarial_Patches_on_Face_Recognition_With_Generative_CVPR_2021_paper.pdf)\n- **WebFace260M**: 揭示百万级深度人脸识别强大能力的基准数据集 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhu_WebFace260M_A_Benchmark_Unveiling_the_Power_of_Million-Scale_Deep_Face_CVPR_2021_paper.pdf)\n- **FaceSec**: 面向人脸识别系统的细粒度鲁棒性评估框架 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FTong_FaceSec_A_Fine-Grained_Robustness_Evaluation_Framework_for_Face_Recognition_Systems_CVPR_2021_paper.pdf)\n- 面向人脸识别的球面置信度学习 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Spherical_Confidence_Learning_for_Face_Recognition_CVPR_2021_paper.pdf)\n- 一致的实例假阳性可提升人脸识别的公平性 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FXu_Consistent_Instance_False_Positive_Improves_Fairness_in_Face_Recognition_CVPR_2021_paper.pdf)\n- **CRFace**: 一种与模型无关的人脸检测精炼置信度排序器 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FVesdapunt_CRFace_Confidence_Ranker_for_Model-Agnostic_Face_Detection_Refinement_CVPR_2021_paper.pdf)\n- **HLA-Face**: 用于低光照下人脸检测的高低联合自适应方法 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FWang_HLA-Face_Joint_High-Low_Adaptation_for_Low_Light_Face_Detection_CVPR_2021_paper.pdf)\n- 基于结构感知的大规模107节点图上的人脸聚类 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FShen_Structure-Aware_Face_Clustering_on_a_Large-Scale_Graph_With_107_Nodes_CVPR_2021_paper.pdf)\n- **img2pose**: 基于6DoF的人脸对齐与检测及面部姿态估计 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FAlbiero_img2pose_Face_Alignment_and_Detection_via_6DoF_Face_Pose_Estimation_CVPR_2021_paper.pdf)\n- **Clusformer**: 一种基于Transformer的聚类方法，用于无监督的大规模人脸与视觉地标识别 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FNguyen_Clusformer_A_Transformer_Based_Clustering_Approach_to_Unsupervised_Large-Scale_Face_CVPR_2021_paper.pdf)\n- 基于自我估计残差年龄嵌入的连续人脸衰老 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Continuous_Face_Aging_via_Self-Estimated_Residual_Age_Embedding_CVPR_2021_paper.pdf)\n- 当年龄不变的人脸识别遇上人脸年龄合成：一个多任务学习框架 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FHuang_When_Age-Invariant_Face_Recognition_Meets_Face_Age_Synthesis_A_Multi-Task_CVPR_2021_paper.pdf)\n- **SDD-FIQA**: 基于相似性分布距离的无监督人脸图像质量评估 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FOu_SDD-FIQA_Unsupervised_Face_Image_Quality_Assessment_With_Similarity_Distribution_Distance_CVPR_2021_paper.pdf)\n- **TediGAN**: 文本引导的多样化人脸图像生成与操控 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FXia_TediGAN_Text-Guided_Diverse_Face_Image_Generation_and_Manipulation_CVPR_2021_paper.pdf)\n- 内嵌GAN先验的网络用于野外盲人脸部修复 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FYang_GAN_Prior_Embedded_Network_for_Blind_Face_Restoration_in_the_CVPR_2021_paper.pdf)\n- 反演生成式对抗渲染器以进行人脸重建 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FPiao_Inverting_Generative_Adversarial_Renderer_for_Face_Reconstruction_CVPR_2021_paper.pdf)\n- 用于盲人脸部修复的渐进式语义感知风格转换 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FChen_Progressive_Semantic-Aware_Style_Transformation_for_Blind_Face_Restoration_CVPR_2021_paper.pdf)\n- 借助生成式面部先验实现面向真实世界的盲人脸部修复 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FWang_Towards_Real-World_Blind_Face_Restoration_With_Generative_Facial_Prior_CVPR_2021_paper.pdf)\n- **FaceInpainter**: 高保真度的人脸跨异构域适应 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_FaceInpainter_High_Fidelity_Face_Adaptation_to_Heterogeneous_Domains_CVPR_2021_paper.pdf)\n- 单张照片实现百万像素级人脸互换 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhu_One_Shot_Face_Swapping_on_Megapixels_CVPR_2021_paper.pdf)\n- 高保真且任意的人脸编辑 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FGao_High-Fidelity_and_Arbitrary_Face_Editing_CVPR_2021_paper.pdf)\n- 寻找声音的形状：一种学习语音-人脸关联的自适应框架 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FWen_Seeking_the_Shape_of_Sound_An_Adaptive_Framework_for_Learning_CVPR_2021_paper.pdf)\n- 基于流引导的一次性高分辨率视听数据集生成会说话的人脸 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhang_Flow-Guided_One-Shot_Talking_Face_Generation_With_a_High-Resolution_Audio-Visual_Dataset_CVPR_2021_paper.pdf)\n- 通过隐式模块化视听表征实现姿态可控的会说话的人脸生成 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhou_Pose-Controllable_Talking_Face_Generation_by_Implicitly_Modularized_Audio-Visual_Representation_CVPR_2021_paper.pdf)\n- 单目重建神经人脸反射场 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FR_Monocular_Reconstruction_of_Neural_Face_Reflectance_Fields_CVPR_2021_paper.pdf)\n- 从图像和视频中学习完整的3D可变形人脸模型 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FR_Learning_Complete_3D_Morphable_Face_Models_From_Images_and_Videos_CVPR_2021_paper.pdf)\n- 通过网络内优化实现可绑定骨骼的3D人脸重建 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FBai_Riggable_3D_Face_Reconstruction_via_In-Network_Optimization_CVPR_2021_paper.pdf)\n- 从野外照片集中学习聚合与个性化3D人脸 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhang_Learning_To_Aggregate_and_Personalize_3D_Face_From_In-the-Wild_Photo_CVPR_2021_paper.pdf)\n- 提升2D StyleGAN以实现3D感知的人脸生成 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FShi_Lifting_2D_StyleGAN_for_3D-Aware_Face_Generation_CVPR_2021_paper.pdf)\n- **3DCaricShop**: 一个用于单视图3D漫画人脸重建的数据集及基线方法 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FQiu_3DCaricShop_A_Dataset_and_a_Baseline_Method_for_Single-View_3D_CVPR_2021_paper.pdf)\n- 幻觉人脸重现 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FSong_Pareidolia_Face_Reenactment_CVPR_2021_paper.pdf)\n- 嘴唇不会说谎：一种可泛化且鲁棒的人脸伪造检测方法 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FHaliassos_Lips_Dont_Lie_A_Generalisable_and_Robust_Approach_To_Face_CVPR_2021_paper.pdf)\n- 空间-相位浅层学习：重新思考频域中的人脸伪造检测 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLiu_Spatial-Phase_Shallow_Learning_Rethinking_Face_Forgery_Detection_in_Frequency_Domain_CVPR_2021_paper.pdf)\n- 频率感知的判别特征学习，由单中心损失监督用于人脸伪造检测 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Frequency-Aware_Discriminative_Feature_Learning_Supervised_by_Single-Center_Loss_for_Face_CVPR_2021_paper.pdf)\n- 利用高频特征泛化人脸伪造检测 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLuo_Generalizing_Face_Forgery_Detection_With_High-Frequency_Features_CVPR_2021_paper.pdf)\n- 基于3D分解的人脸伪造检测 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhu_Face_Forgery_Detection_by_3D_Decomposition_CVPR_2021_paper.pdf)\n- 探索人脸流形上的对抗性虚假图像 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FLi_Exploring_Adversarial_Fake_Images_on_Face_Manifold_CVPR_2021_paper.pdf)\n- 代表性伪造挖掘用于假人脸检测 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FWang_Representative_Forgery_Mining_for_Fake_Face_Detection_CVPR_2021_paper.pdf)\n- 用于RGBD人脸防欺骗的跨模态焦点损失 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FGeorge_Cross_Modal_Focal_Loss_for_RGBD_Face_Anti-Spoofing_CVPR_2021_paper.pdf)\n- 通过深度光照适应实现AR\u002FVR中的高保真人脸追踪 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FChen_High-Fidelity_Face_Tracking_for_ARVR_via_Deep_Lighting_Adaptation_CVPR_2021_paper.pdf)\n- 朝着具有逼真阴影的高保真人脸重照明迈进 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FHou_Towards_High_Fidelity_Face_Relighting_With_Realistic_Shadows_CVPR_2021_paper.pdf)\n- **IronMask**: 用于保护深度人脸模板的模块化架构 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FKim_IronMask_Modular_Architecture_for_Protecting_Deep_Face_Template_CVPR_2021_paper.pdf)\n- 野外的人脸取证 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FCVPR2021\u002Fpapers\u002FZhou_Face_Forensics_in_the_Wild_CVPR_2021_paper.pdf)\n\n**ICCV2021**  \n- 基于嵌入和头部钩子的全身-人脸联合检测 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FWan_Body-Face_Joint_Detection_via_Embedding_and_Head_Hook_ICCV_2021_paper.pdf)  \n- 面向深度人脸识别的元监督自适应标签噪声清理 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FZhang_Adaptive_Label_Noise_Cleaning_With_Meta-Supervision_for_Deep_Face_Recognition_ICCV_2021_paper.pdf)  \n- 教师-学生对抗性深度幻觉用于提升人脸识别性能 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FUppal_Teacher-Student_Adversarial_Depth_Hallucination_To_Improve_Face_Recognition_ICCV_2021_paper.pdf)  \n- **DAM**：用于人脸识别的差异对齐度量 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FLiu_DAM_Discrepancy_Alignment_Metric_for_Face_Recognition_ICCV_2021_paper.pdf)  \n- **SynFace**：基于合成数据的人脸识别 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FQiu_SynFace_Face_Recognition_With_Synthetic_Data_ICCV_2021_paper.pdf)  \n- **PASS**：用于缓解人脸识别中偏见的受保护属性抑制系统 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FDhar_PASS_Protected_Attribute_Suppression_System_for_Mitigating_Bias_in_Face_ICCV_2021_paper.pdf)  \n- 无年龄变化的人脸识别中的解耦表征：互信息最小化视角 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FHou_Disentangled_Representation_for_Age-Invariant_Face_Recognition_A_Mutual_Information_Minimization_ICCV_2021_paper.pdf)  \n- 基于人脸循环一致性的面部表征学习 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FChang_Learning_Facial_Representations_From_the_Cycle-Consistency_of_Face_ICCV_2021_paper.pdf)  \n- 通过解耦身份信息操纵实现个性化且可逆的人脸去标识化 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FCao_Personalized_and_Invertible_Face_De-Identification_by_Disentangled_Identity_Information_Manipulation_ICCV_2021_paper.pdf)  \n- **ADNet**：利用误差偏差朝法线方向进行人脸对齐 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FHuang_ADNet_Leveraging_Error-Bias_Towards_Normal_Direction_in_Face_Alignment_ICCV_2021_paper.pdf)  \n- 通过生成对抗性身份掩码实现人脸加密 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FYang_Towards_Face_Encryption_by_Generating_Adversarial_Identity_Masks_ICCV_2021_paper.pdf)  \n- 用于图像和视频中解耦人脸编辑的潜在变换器 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FYao_A_Latent_Transformer_for_Disentangled_Face_Editing_in_Images_and_ICCV_2021_paper.pdf)  \n- **Re-Aging GAN**：迈向个性化的人脸年龄转换 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FMakhmudkhujaev_Re-Aging_GAN_Toward_Personalized_Face_Age_Transformation_ICCV_2021_paper.pdf)  \n- 解耦寿命阶段的人脸合成 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FHe_Disentangled_Lifespan_Face_Synthesis_ICCV_2021_paper.pdf)  \n- 基于属性操纵的人脸图像检索 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FZaeemzadeh_Face_Image_Retrieval_With_Attribute_Manipulation_ICCV_2021_paper.pdf)  \n- 基于条件估计的自监督3D人脸重建 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FWen_Self-Supervised_3D_Face_Reconstruction_via_Conditional_Estimation_ICCV_2021_paper.pdf)  \n- 基于自监督学习和光线追踪实现高保真单目人脸重建，支持丰富的反射特性 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FDib_Towards_High_Fidelity_Monocular_Face_Reconstruction_With_Rich_Reflectance_Using_ICCV_2021_paper.pdf)  \n- 基于条件估计的自监督3D人脸重建 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FWen_Self-Supervised_3D_Face_Reconstruction_via_Conditional_Estimation_ICCV_2021_paper.pdf)  \n- 基于体积采样的拓扑一致性多视角人脸推断 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FLi_Topologically_Consistent_Multi-View_Face_Inference_Using_Volumetric_Sampling_ICCV_2021_paper.pdf)  \n- **假装直到成功**：仅使用合成数据进行野外人脸分析 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FWood_Fake_It_Till_You_Make_It_Face_Analysis_in_the_ICCV_2021_paper.pdf)  \n- 探索时间一致性以实现更通用的视频人脸伪造检测 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FZheng_Exploring_Temporal_Coherence_for_More_General_Video_Face_Forgery_Detection_ICCV_2021_paper.pdf)  \n- **OpenForensics**：用于野外多人脸伪造检测与分割的大规模挑战性数据集 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FLe_OpenForensics_Large-Scale_Challenging_Dataset_for_Multi-Face_Forgery_Detection_and_Segmentation_ICCV_2021_paper.pdf)  \n- 新型人脸呈现攻击的检测与持续学习 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FRostami_Detection_and_Continual_Learning_of_Novel_Face_Presentation_Attacks_ICCV_2021_paper.pdf)  \n- **MeshTalk**：利用跨模态解耦技术从语音驱动3D人脸动画 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FRichard_MeshTalk_3D_Face_Animation_From_Speech_Using_Cross-Modality_Disentanglement_ICCV_2021_paper.pdf)  \n- 通过窥视一次性样例超分辨率跨域人脸缩略图 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FLi_Super-Resolving_Cross-Domain_Face_Miniatures_by_Peeking_at_One-Shot_Exemplar_ICCV_2021_paper.pdf)  \n- 多模态联想桥接与记忆：从人脸视频中回忆起语音声学特征 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FKim_Multi-Modality_Associative_Bridging_Through_Memory_Speech_Sound_Recollected_From_Face_ICCV_2021_paper.pdf)  \n- **FACIAL**：通过隐式属性学习合成动态说话人脸 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FZhang_FACIAL_Synthesizing_Dynamic_Talking_Face_With_Implicit_Attribute_Learning_ICCV_2021_paper.pdf)  \n- **VariTex**：变分神经网络人脸纹理 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FBuhler_VariTex_Variational_Neural_Face_Textures_ICCV_2021_paper.pdf)  \n- 在缺乏完整人脸纹理的情况下学习高保真的人脸纹理补全 [[论文]](https:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent\u002FICCV2021\u002Fpapers\u002FKim_Learning_High-Fidelity_Face_Texture_Completion_Without_Complete_Face_Texture_ICCV_2021_paper.pdf)\n\n###### 2020-10-02\n**ICIP2020**  \n- 用于动态人脸识别的3D稀疏形变特征\n- 基于异常检测的客户端特定人脸欺骗检测堆叠集成方法\n- 用于多人脸跟踪的自适应聚合轨迹链接方法\n- 用于人脸合成与姿态不变人脸识别的注意力选择网络\n- **EDGE-GAN**: 边缘条件下的多视角人脸图像生成\n- 提取深度局部特征以检测篡改的人脸图像\n- 基于灰度编码光场的人脸认证\n- 基于深度特征重建网络的低光照下人脸识别\n- 利用GAN进行判别性特征恢复以提升退化人脸的检测与识别\n- **QAMFACE**: 用于人脸识别的二次加性角度间隔损失\n- 具有几何感知特征变换的真实感说话人脸合成\n- 深度人脸识别中的三元组蒸馏  \n\n**ECCV2020**\n- **“看，没有地标点！”** – 无监督、基于模型的密集人脸对齐\n- 通过解耦潜在特征的分层人脸衰老\n- 用于浅层人脸学习的半暹罗训练\n- 基于3D人脸先验指导的人脸超分辨率\n- 个性化人脸建模以改善人脸重建和运动重定向\n- **ProgressFace**: 面向尺度感知的渐进式人脸检测学习\n- 基于人类材质感知的人脸反欺骗\n- **超越3DMM空间**: 向精细化3D人脸重建迈进\n- 基于深度多尺度组件字典的盲人像修复\n- 带有不等式约束且鲁棒的3D人脸模型拟合\n- **BroadFace**: 一次性观察数万人进行人脸识别\n- 可解释的人脸识别\n- **CONFIG**: 可控的神经网络人脸图像生成\n- **子中心ArcFace**: 通过大规模噪声网络人脸数据提升人脸识别性能\n- **CelebA-Spoof**: 大规模、标注丰富的面部反欺骗数据集\n- **频率思维**: 通过挖掘频率感知线索进行人脸伪造检测\n- 面向人脸分割的边缘感知图表示学习与推理\n- 学习基于流的特征扭曲以实现人脸正面化，并采用光照不一致的监督\n- **CAFE-GAN**: 借助互补注意力特征实现任意人脸属性编辑\n- 自监督单目3D人脸重建：基于遮挡感知的多视角几何一致性\n- **生成适应**: 用于监控场景下人脸识别的分辨率自适应网络\n- **标题监督的人脸识别**: 在无需人工标注的情况下训练最先进的人脸模型\n- 通用对抗补丁在人脸检测中的设计与解读\n- **JNR**: 基于关节的神经绑定表示法，用于紧凑的3D人脸建模\n- 关于解耦通用人脸反欺骗中的欺骗痕迹\n- 向人脸分析算法偏差的因果基准测试迈进\n- 向快速、准确、稳定的3D密集人脸对齐迈进\n- 基于解耦表征学习的人脸反欺骗\n- 学习预测显著人脸：一种新型视听显著性模型\n- **MEAD**: 用于情感化说话人脸生成的大规模视听数据集\n- **Margin-Mix**: 用于表情识别的半监督学习\n- 基于密码条件的身份匿名化与去匿名化：使用人脸身份变换器\n- 通过聚类野外未标记人脸来提升人脸识别性能\n- 基于排他性-一致性正则化的知识蒸馏用于人脸识别\n- **BioMetricNet**: 通过学习正则化到高斯分布的度量，实现深度非约束条件下的人脸验证\n- 从单张人脸图像重建眼镜的3D形状\n- 基于残差型跨物种等变网络的深度跨物种特征学习，用于动物人脸识别\n- 高分辨率零样本域适应：针对合成渲染人脸图像\n- **ByeGlassesGAN**: 保留身份信息的眼镜去除技术，适用于人脸图像\n- 联合去偏见：人脸识别与人口统计属性估计\n- 借助主干-分支生成对抗网络合成耦合的3D人脸模态\n- 通过分布蒸馏损失提升难样本下的人脸识别性能\n- 流形投影用于人脸识别的对抗防御  \n\n**SIGGRAPH2020**\n- 用于高效3D打印定格动画人脸的系统\n- 基于深度微分子空间重建的精确人脸绑定近似\n- **DeepFaceDrawing**: 基于草图的深度人脸图像生成\n- **眼睛说了算**: 一个集成眼睑与人脸模型的逼真面部动画系统  \n\n**IJCAI2020**  \n- 用于遮挡不变人脸识别的偏置特征学习\n- 基于参考的人脸部件编辑\n- 借助注意力驱动的视听一致性学习实现任意说话人脸的生成\n- **FakeSpotter**: 一种简单而鲁棒的基线方法，用于检测AI合成的假人脸\n\n**CVPR2020**  \n- 带可关闭跳跃连接的跨模态深度人脸法线  \n- 面部生成的一次性域适应  \n- 通过一致性学习结构以实现人脸分割与解析  \n- **BFBox**：为面部检测器搜索适合的骨干网络和特征金字塔网络  \n- **领域平衡**：长尾分布下的人脸识别  \n- **FReeNet**：多身份人脸重演  \n- 学习跨身份人脸重演中的身份不变运动表示  \n- **全局-局部GCN**：大规模标签噪声清理用于人脸识别  \n- **3FabRec**：基于重建的快速少样本人脸对齐  \n- 面向野外场景的假脸检测中的全局纹理增强  \n- **CurricularFace**：深度人脸识别的自适应课程学习损失  \n- 关于数字人脸操纵的检测  \n- 面部防伪中的深度空间梯度与时间深度学习  \n- **ReDA**：用于3D人脸重建的强化可微属性  \n- 基于多领域解耦表征学习的跨领域人脸呈现攻击检测  \n- **FaceScape**：大规模高质量3D人脸数据集及详细的可绑定3D人脸预测  \n- 解释GAN的潜在空间以进行语义化人脸编辑  \n- **旋转并渲染**：从单视角图像无监督地实现照片级真实感人脸旋转  \n- 基于3D模仿对比学习的解耦可控人脸图像生成  \n- 面向聚类的密度感知特征嵌入  \n- 学习“听觉”以提升人脸超分辨率  \n- 学习基于物理的人脸属性形成  \n- **LUVLi人脸对齐**：估计关键点位置、不确定性及可见性概率  \n- 在未见领域中学习元人脸识别  \n- 通过解耦独立因子实现跨光谱人脸幻化  \n- 基于注意力恢复与关键点估计迭代协作的深度人脸超分辨率  \n- 人脸识别中的数据不确定性学习  \n- 面部X光用于更通用的伪造人脸检测  \n- **Vec2Face**：从人脸识别的黑盒特征中揭示人脸  \n- **FM2u-Net**：用于妆容无关人脸验证的面部形态多分支网络  \n- 不确定性感知网格解码器用于高保真3D人脸重建  \n- 结合多参考图像与自适应空间特征融合的增强型盲人脸修复  \n- **SER-FIQ**：基于随机嵌入鲁棒性的无监督人脸图像质量估计  \n- 利用图卷积网络从野外图像中实现高保真3D人脸重建  \n- **RDCFace**：用于人脸识别的径向畸变校正  \n- 搜索用于面部防伪的中心差分卷积网络  \n- **RetinaFace**：在野外场景中的一次性多层级人脸定位  \n- 使用偏度感知强化学习缓解人脸识别中的偏差  \n- **DeeperForensics-1.0**：用于现实世界伪造人脸检测的大规模数据集  \n- **GroupFace**：学习潜在群体并构建基于群体的人脸表征  \n- 可形变人脸反照率模型  \n- 学习Oracle注意力以实现高保真人脸补全  \n- 学习在定向光照下由物理引导的人脸重打光  \n- 向面向深度人脸识别的通用表征学习迈进  \n- 适用于高效低比特人脸识别的旋转一致边界损失  \n- **HAMBox**：深入挖掘用于面部检测的高质量锚框  \n- 用于人脸识别的层次化金字塔多样注意力网络  \n- 基于强化学习的动态人脸视频分割  \n- 复制粘贴GAN：从阴影缩略图中幻化人脸  \n- 单侧域泛化用于面部防伪  \n\n**AAAI2020**  \n- 用于游戏角色自动创建的快速稳健的人脸到参数转换  \n- 面向人脸识别的误分类向量引导软最大值损失  \n- 学习零样本和少样本面部防伪的元模型  \n- 通过草图合成学习去模糊人脸图像  \n- **FAN-Face**：对深度人脸识别的简单正交改进  \n- **KPNet**：迈向极简面部检测器  \n- 朝着大规模无标注视频的全监督人脸对齐迈进  \n- 正则化的细粒度元面部防伪  \n- **GDFace**：用于多视角人脸图像合成的门控变形  \n- 基于自我监督的身份与姿态解耦实现逼真的人脸重演  \n- **MarioNETte**：保留未见目标身份的少样本人脸重演  \n- 一个新的数据集和边界注意力语义分割用于人脸解析  \n- 基于运动自适应反馈单元的视频人脸超分辨率  \n- 面部属性胶囊用于噪声环境下的人脸超分辨率  \n- 微小人脸的联合超分辨率与对齐  \n###### 2020年1月26日  \n- **UGG**：用于非约束视频人脸识别的轨迹片段间上下文关联不确定性建模  \n- **PDSN**：基于成对差异暹罗网络的掩码学习实现遮挡鲁棒的人脸识别  \n- 用于精准人脸识别的注意力特征对关系网络  \n- **PFE**：概率人脸嵌入  \n- 向可解释的人脸识别迈进  \n- **协同挖掘**：带有噪声标签的深度人脸识别  \n- **公平损失**：面向深度人脸识别的边界感知强化学习  \n- 用于集合基础人脸识别的判别式学习凸模型  \n- **DVG**：用于低样本异质性人脸识别的双重变分生成  \n- **CDP**：面向人脸识别的大规模无标注数据中的共识驱动传播  \n\n- **BCL**：未知簇数的视频人脸聚类  \n\n- **DeCaFA**：用于野外场景中人脸对齐的深度卷积级联  \n- **AWing**：基于热图回归的自适应翼损失，用于鲁棒的人脸对齐  \n- **KDN**：基于核密度深度神经网络的人脸对齐  \n\n- **DF2Net**：用于精细3D人脸重建的密集-精细-更精细网络  \n- 使用3D人脸先验知识进行视频人脸去模糊  \n- 半监督单目3D人脸重建，实现端到端形状保持的域迁移  \n- **3DFC**：从多样化原始扫描数据中进行3D人脸建模  \n\n- 视频中实时人脸去标识化  \n- 用于游戏角色自动创建的人脸到参数转换  \n- **SC-FEGAN**：结合用户草图与色彩的人脸编辑生成对抗网络  \n- **FSGAN**：主体无关的人脸交换与重演  \n- **Make a Face**：迈向任意高保真度的人脸操控  \n- 使用3D可形变模型和生成对抗网络进行人脸去遮挡  \n- **FRV**：利用生成对抗网络从语音重建人脸  \n- **从推理到生成**：端到端完全自监督的人脸从语音生成  \n- **PFSR**：基于面部关键点注意力的渐进式人脸超分辨率\n\n###### 2019-07-11\n- 基于不完美人脸数据的深度人脸识别\n- 面向长尾噪声数据的不均衡训练深度人脸识别\n- **RegularFace**：通过独占正则化实现深度人脸识别\n- **UniformFace**：学习用于人脸识别的深度等分布表征\n- **P2SGrad**：优化深度人脸模型的精细化梯度\n- **AdaptiveFace**：用于人脸识别的自适应间隔和采样策略\n- **AdaCos**：自适应缩放余弦 logits，以有效学习深度人脸表征\n- 用于鲁棒人脸识别的低秩拉普拉斯-均匀混合模型\n- **NoiseFace**：一种容忍噪声的人脸识别 CNN 训练范式\n- 针对代表性不足数据的人脸识别特征迁移学习\n- **Led3D**：一种轻量高效、用于识别低质量 3D 人脸的深度方法\n- 用于跨模型人脸识别的 R3 对抗网络\n\n- **RetinaFace**：野外单阶段密集人脸定位\n- 用于尺度不变人脸检测的分组采样\n- **FA-RPN**：用于人脸检测的浮动区域建议框\n\n- **语义对齐**：为面部关键点检测寻找语义一致的真值\n- 通过遮挡自适应深度网络实现鲁棒的面部关键点检测\n\n- **LTC**：在亲和图上学习聚类人脸\n\n- **FECNet**：一种用于面部表情相似性的小型嵌入模型\n- **LBVCNN**：用于从图像序列中识别面部表情的局部二值体积卷积神经网络\n\n- 面部动作单元强度估计的联合表征与估计器学习\n- 结合个体特定形状正则化的局部关系学习，用于面部动作单元检测\n- **TCAE**：基于视频的自监督表征学习，用于面部动作单元检测\n- **JAANet**：用于联合面部动作单元检测与人脸对齐的深度自适应注意力机制\n\n- **2DASL**：基于 2D 辅助自监督学习，从单张图像联合重建 3D 人脸并进行密集人脸对齐\n- **MVF-Net**：多视角 3D 人脸可变形模型回归\n- 超过 2500 FPS 的密集 3D 人脸解码：联合纹理与形状的卷积网格解码器\n- 朝向高保真非线性 3D 人脸可变形模型迈进\n- 结合 3D 可变形模型：大规模人脸与头部模型\n- 用于 3D 人脸形状的解耦表征学习\n- 高保真人脸模型的自监督适配，用于单目姿态跟踪\n- **MMFace**：一种用于无约束人脸重建的多指标回归网络\n- 在无 3D 监督的情况下，从图像中回归 3D 人脸形状与表情\n- 提升局部形状匹配，用于密集 3D 人脸对应\n- **FML**：从视频中学习人脸模型\n- **2DASL**：基于 2D 辅助自监督学习，从单张图像联合重建 3D 人脸并进行密集人脸对齐\n\n- **ATVGnet**：具有动态像素级损失的层次化跨模态说话人脸生成\n- **Speech2Face**：学习声音背后的面孔\n\n- 野外极端姿态与表情下的无监督人脸归一化\n- **GANFIT**：用于高保真 3D 人脸重建的生成对抗网络拟合\n\n- **BeautyGAN**：基于深度生成对抗网络的实例级面部化妆迁移\n- **FUNIT**：少样本无监督图像到图像转换\n- 基于深度强化学习的视频中自动人脸老化\n- 基于小波的生成对抗网络实现属性感知的人脸老化\n- **SAGAN**：带有空间注意力的生成对抗网络，用于人脸属性编辑\n- **APDrawingGAN**：利用层次化 GAN 从人脸照片生成艺术肖像画\n- **StyleGAN**：一种基于风格的生成器架构，适用于生成对抗网络\n\n- 3D 引导的细粒度人脸操控\n- **SemanticComponent**：用于人脸属性操控的语义组件分解\n\n- **数据集与基准**：大规模多模态人脸防欺骗的数据集与基准\n- 零样本人脸防欺骗的深度树学习\n\n- 用于年龄不变人脸识别的去相关对抗学习\n- 多对抗判别式深度领域泛化，用于人脸呈现攻击检测\n- 面向人脸识别的高效决策型黑盒对抗攻击\n\n- **Speech2Face**：学习声音背后的面孔\n- **JFDFMR**：用于多人脸的联合人脸检测与面部运动重定向\n- **ATVGnet**：具有动态像素级损失的层次化跨模态说话人脸生成\n\n- 使用解剖肌肉进行高质量人脸捕捉\n- 单目全身捕捉：在野外捕捉面部、身体和双手的姿态\n- 富有表现力的身体捕捉：从一张图像中同时获取 3D 手、脸和身体\n\n###### 2019-04-06\n- **ISRN**：改进的选择性精炼网络，用于人脸检测\n- **DSFD**：双击人脸检测器\n- **PyramidBox++**：高性能微小人脸检测器\n- **VIM-FD**：鲁棒且高性能的人脸检测器\n- **SHF**：通过在困难图像上学习小人脸实现鲁棒的人脸检测\n- **SRN**：用于高性能人脸检测的选择性精炼网络\n- **SFDet**：单次拍摄、具备尺度感知能力的实时人脸检测网络\n- **JFDFMR**：用于多人脸的联合人脸检测与面部运动重定向\n- **PFLD**：一款实用的面部关键点检测器\n- **LinkageFace**：基于图卷积网络的链接式人脸聚类\n- **MLT**：人脸识别：一项基于新型多层级分类体系的综述\n- **GhostVLAD**：用于集合式人脸识别的 GhostVLAD\n- **DocFace+**：身份证件与自拍匹配\n- **DiF**：人脸多样性\n- **2018 年综述**：人脸识别：从传统方法到深度学习方法\n\n###### 2019-01-12\n- **2018 年综述**：深度面部表情识别：一项综述\n- **2018 年综述**：深度人脸识别：一项综述\n- **SphereFace+(MHE)**：面向最小超球面能量的学习\n- **HyperFace**：一个用于人脸检测、关键点定位、姿态估计和性别识别的深度多任务学习框架\n\n###### 2018-12-01\n- **FRVT**: 人脸识别供应商测试\n- **GANimation**: 基于单张图像的解剖学感知人脸动画\n- **StarGAN**: 用于多域图像到图像转换的统一生成对抗网络\n- **Faceswap**: 一种利用深度学习识别并交换图片和视频中人脸的工具\n- **HF-PIM**: 学习高保真姿态不变模型以实现高分辨率人脸正面化\n- **PRNet**: 结合位置图回归网络的联合3D人脸重建与密集对齐\n- **LAB**: 关注边界：一种边界感知的人脸对齐算法\n- **Super-FAN**: 利用GAN集成人脸关键点定位与任意姿态下真实世界低分辨率人脸的超分辨率重建\n- **Face-Alignment**: 我们距离解决2D和3D人脸对齐问题还有多远？（附带包含23万个3D人脸关键点的数据集）\n- **Face3D**: 用于处理3D人脸的Python工具\n- **IMDb-Face**: 人脸识别的魔鬼藏在噪声中\n- **AAM-Softmax(CCL)**: 通过集中式坐标学习进行人脸识别\n- **AM-Softmax**: 用于人脸验证的加性间隔Softmax\n- **FeatureIncay**: 用于表征正则化的Feature Incay\n- **NormFace**: 用于人脸验证的L2超球面嵌入\n- **CocoLoss**: 重新思考大规模识别中的特征判别与聚合\n- **L-Softmax**: 用于卷积神经网络的大间隔Softmax损失\n\n###### 2018-07-21\n- **MobileFace**: 移动设备上的人脸识别解决方案\n- **Trillion Pairs**: 挑战3：人脸特征测试\u002F万亿对\n- **MobileFaceNets**: 用于在移动设备上实现准确实时人脸验证的高效CNN\n\n###### 2018-04-20\n- **PyramidBox**: 上下文辅助的单次检测型人脸检测器\n- **PCN**: 基于渐进校准网络的实时旋转不变人脸检测\n- **S³FD**: 单次检测的尺度不变人脸检测器\n- **SSH**: 单阶段无头部人脸检测器\n- **NPD**: 一种快速且精确的无约束人脸检测器\n- **PICO**: 基于决策树组织的像素强度比较的目标检测\n- **libfacedetection**: 一个用于图像中人脸检测和人脸关键点检测的快速二进制库。\n- **SeetaFaceEngine**: 包括SeetaFace检测、SeetaFace对齐和SeetaFace识别功能。\n- **FaceID**: 使用RGBD图像上的人脸嵌入和孪生网络实现的iPhone X Face ID。\n\n###### 2018-03-28\n- **InsightFace(ArcFace)**: 2D和3D人脸分析项目\n- **CosFace**: 用于深度人脸识别的大间隔余弦损失\n\n## 🔖 人脸基准测试与数据集\n#### 人脸识别数据\n- **DiF**: 多样性人脸[[项目]](https:\u002F\u002Fwww.research.ibm.com\u002Fartificial-intelligence\u002Ftrusted-ai\u002Fdiversity-in-faces\u002F) [[博客]](https:\u002F\u002Fwww.ibm.com\u002Fblogs\u002Fresearch\u002F2019\u002F01\u002Fdiversity-in-faces\u002F)\n- **FRVT**: 人脸识别供应商测试[[项目]](https:\u002F\u002Fwww.nist.gov\u002Fprograms-projects\u002Fface-recognition-vendor-test-frvt) [[排行榜]](https:\u002F\u002Fwww.nist.gov\u002Fprograms-projects\u002Fface-recognition-vendor-test-frvt-ongoing)\n- **IMDb-Face**: 人脸识别的魔鬼藏在噪声中(**5.9万人，170万张图像**) [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ECCV_2018\u002Fpapers\u002FLiren_Chen_The_Devil_of_ECCV_2018_paper.pdf \"ECCV2018\") [[数据集]](https:\u002F\u002Fgithub.com\u002Ffwang91\u002FIMDb-Face)\n- **Trillion Pairs**: 挑战3：人脸特征测试\u002F万亿对数据(**MS-Celeb-1M-v1c，包含86,876个ID、3,923,399张对齐图像 + Asian-Celeb，包含93,979个ID、2,830,146张对齐图像**) [[基准测试]](http:\u002F\u002Ftrillionpairs.deepglint.com\u002Foverview \"DeepGlint\") [[数据集]](http:\u002F\u002Ftrillionpairs.deepglint.com\u002Fdata) [[结果]](http:\u002F\u002Ftrillionpairs.deepglint.com\u002Fresults)\n- **MF2**: 百万级人脸识别的公平竞争平台(**67.2万人，470万张图像**) [[论文]](https:\u002F\u002Fhomes.cs.washington.edu\u002F~kemelmi\u002Fms.pdf \"CVPR2017\") [[数据集]](http:\u002F\u002Fmegaface.cs.washington.edu\u002Fdataset\u002Fdownload_training.html) [[结果]](http:\u002F\u002Fmegaface.cs.washington.edu\u002Fresults\u002Ffacescrub_challenge2.html) [[基准测试]](http:\u002F\u002Fmegaface.cs.washington.edu\u002F)\n- **MegaFace**: MegaFace基准测试：大规模识别的100万张人脸(**69万人，100万张图像**) [[论文]](http:\u002F\u002Fmegaface.cs.washington.edu\u002FKemelmacherMegaFaceCVPR16.pdf \"CVPR2016\") [[数据集]](http:\u002F\u002Fmegaface.cs.washington.edu\u002Fparticipate\u002Fchallenge.html) [[结果]](http:\u002F\u002Fmegaface.cs.washington.edu\u002Fresults\u002Ffacescrub.html) [[基准测试]](http:\u002F\u002Fmegaface.cs.washington.edu\u002F)\n- **UMDFaces**: 用于训练深度网络的标注人脸数据集(**8千人，36.7万张图像，包含姿态、21个关键点和性别信息**) [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1611.01484.pdf \"arXiv2016\") [[数据集]](http:\u002F\u002Fwww.umdfaces.io\u002F)\n- **MS-Celeb-1M**: 大规模人脸识别的数据集与基准测试(**10万人，1000万张图像**) [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1607.08221.pdf \"ECCV2016\") [[数据集]](http:\u002F\u002Fwww.msceleb.org\u002Fdownload\u002Fsampleset) [[结果]](http:\u002F\u002Fwww.msceleb.org\u002Fleaderboard\u002Ficcvworkshop-c1) [[基准测试]](http:\u002F\u002Fwww.msceleb.org\u002F) [[项目]](https:\u002F\u002Fwww.microsoft.com\u002Fen-us\u002Fresearch\u002Fproject\u002Fms-celeb-1m-challenge-recognizing-one-million-celebrities-real-world\u002F)\n- **VGGFace2**: 用于跨姿态和年龄识别人脸的数据集(**9千人，330万张图像**) [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1710.08092.pdf \"arXiv2017\") [[数据集]](http:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vgg\u002Fdata\u002Fvgg_face2\u002F)\n- **VGGFace**: 深度人脸识别(**2.6千人，2.6百万张图像**) [[论文]](http:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vgg\u002Fpublications\u002F2015\u002FParkhi15\u002Fparkhi15.pdf \"BMVC2015\") [[数据集]](http:\u002F\u002Fwww.robots.ox.ac.uk\u002F~vgg\u002Fdata\u002Fvgg_face\u002F)\n- **CASIA-WebFace**: 从零开始学习人脸表示(**1万人，50万张图像**) [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1411.7923.pdf \"arXiv2014\") [[数据集]](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fenglish\u002FCASIA-WebFace-Database.html)\n- **LFW**: 野外带标签的人脸：用于研究非约束环境下人脸识别的数据库(**5.7万人，1.3万张图像**) [[报告]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Flfw\u002Flfw.pdf \"UMASS2007\") [[数据集]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Flfw\u002F#download) [[结果]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Flfw\u002Fresults.html) [[基准测试]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Flfw\u002F)\n\n#### 人脸检测数据\n- **WiderFace**: WIDER FACE：人脸检测基准测试(**40万人，3.2万张图像，尺度、姿态和遮挡变化极大**) [[论文]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2016\u002Fpapers\u002FYang_WIDER_FACE_A_CVPR_2016_paper.pdf \"CVPR2016\") [[数据集]](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FWIDERFace\u002F) [[结果]](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FWIDERFace\u002FWiderFace_Results.html) [[基准测试]](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FWIDERFace\u002F)\n- **FDDB**: 非约束环境下人脸检测的基准测试(**5千张人脸，2.8千张图像**) [[报告]](https:\u002F\u002Fpeople.cs.umass.edu\u002F~elm\u002Fpapers\u002Ffddb.pdf \"UMASS2010\") [[数据集]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Ffddb\u002Findex.html#download) [[结果]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Ffddb\u002Fresults.html) [[基准测试]](http:\u002F\u002Fvis-www.cs.umass.edu\u002Ffddb\u002F)\n\n#### 人脸关键点数据\n- **LS3D-W**: 一个大规模3D人脸对齐数据集，通过使用自动化方法以一致的方式为AFLW、300VW、300W和FDDB中的图像标注68个关键点而构建[[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2017\u002Fpapers\u002FBulat_How_Far_Are_ICCV_2017_paper.pdf \"ICCV2017\") [[数据集]](https:\u002F\u002Fadrianbulat.com\u002Fface-alignment)\n- **AFLW**: 野外标注的人脸关键点：用于人脸关键点定位的大规模真实世界数据库(**2.5万张人脸，每张包含21个关键点**) [[论文]](https:\u002F\u002Ffiles.icg.tugraz.at\u002Fseafhttp\u002Ffiles\u002F460c7623-c919-4d35-b24e-6abaeacb6f31\u002Fkoestinger_befit_11.pdf \"BeFIT2011\") [[基准测试]](https:\u002F\u002Fwww.tugraz.at\u002Finstitute\u002Ficg\u002Fresearch\u002Fteam-bischof\u002Flrs\u002Fdownloads\u002Faflw\u002F)\n\n#### 人脸属性数据\n- **CelebA**: 野外深度学习人脸属性(**1万人，20.2万张图像，每张图像包含5个关键点和40个二值属性**) [[论文]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_iccv_2015\u002Fpapers\u002FLiu_Deep_Learning_Face_ICCV_2015_paper.pdf \"ICCV2015\") [[数据集]](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FCelebA.html)\n\n## 🔖 人脸识别\n- 视频中的实时人脸去标识化 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1911.08348 \"ICCV2019\")\n- **UGG**：用于无约束视频人脸识别的轨迹片段间上下文关联不确定性建模 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.02756 \"ICCV2019\")\n- **PDSN**：基于成对差异暹罗网络的掩码学习的遮挡鲁棒人脸识别 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.06290 \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FlinserSnow\u002FPDSN \"PyTorch\")\n- 用于精确人脸识别的注意力特征对关系网络 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.06255 \"ICCV2019\")\n- 概率人脸嵌入 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.09658 \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FseasonSH\u002FProbabilistic-Face-Embeddings \"TensorFlow\")\n- 向可解释的人脸识别迈进 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1805.00611 \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fyubangji123\u002FInterpret_FR \"TensorFlow\") [[项目]](http:\u002F\u002Fcvlab.cse.msu.edu\u002Fproject-interpret-FR)\n- **Co-Mining**：带噪声标签的深度人脸识别 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FWang_Co-Mining_Deep_Face_Recognition_With_Noisy_Labels_ICCV_2019_paper.pdf \"ICCV2019\")\n- **Fair Loss**：面向深度人脸识别的边缘感知强化学习 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FLiu_Fair_Loss_Margin-Aware_Reinforcement_Learning_for_Deep_Face_Recognition_ICCV_2019_paper.pdf \"ICCV2019\")\n- 面向集合型人脸识别的判别式凸模型学习 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FCevikalp_Discriminatively_Learned_Convex_Models_for_Set_Based_Face_Recognition_ICCV_2019_paper.pdf \"ICCV2019\")\n- **DVG**：用于低样本异构人脸识别的双变分生成 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.10203 \"NeurIPS2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FBradyFU\u002FDVG \"PyTorch\")\n- 使用不完美人脸数据的深度人脸识别 [[论文]](https:\u002F\u002Fwww.sciencedirect.com\u002Fscience\u002Farticle\u002Fpii\u002FS0167739X18331133 \"FGCS2019\")\n- 面向长尾噪声数据的深度人脸识别不均衡训练 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhong_Unequal-Training_for_Deep_Face_Recognition_With_Long-Tailed_Noisy_Data_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fzhongyy\u002FUnequal-Training-for-Deep-Face-Recognition-with-Long-Tailed-Noisy-Data \"MXNet\")\n- **RegularFace**：通过独占正则化实现深度人脸识别 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhao_RegularFace_Deep_Face_Recognition_via_Exclusive_Regularization_CVPR_2019_paper.pdf \"CVPR2019\")\n- **UniformFace**：学习用于人脸识别的深度等分布表示 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FDuan_UniformFace_Learning_Deep_Equidistributed_Representation_for_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\")\n- **P2SGrad**：优化深度人脸模型的精细化梯度 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhang_P2SGrad_Refined_Gradients_for_Optimizing_Deep_Face_Models_CVPR_2019_paper.pdf \"CVPR2019\")\n- **AdaptiveFace**：人脸识别中的自适应边距与采样 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FLiu_AdaptiveFace_Adaptive_Margin_and_Sampling_for_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\")\n- **AdaCos**：自适应缩放余弦 logits 以有效学习深度人脸表征 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhang_AdaCos_Adaptively_Scaling_Cosine_Logits_for_Effectively_Learning_Deep_Face_CVPR_2019_paper.pdf \"CVPR2019\") [[代码1]](https:\u002F\u002Fgithub.com\u002Fxialuxi\u002Farcface-caffe \"Caffe\") [[代码2]](https:\u002F\u002Fgithub.com\u002F4uiiurz1\u002Fpytorch-adacos \"PyTorch\")\n- 用于鲁棒人脸识别的低秩拉普拉斯-均匀混合模型 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FDong_Low-Rank_Laplacian-Uniform_Mixed_Model_for_Robust_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\")\n- **NoiseFace**：训练人脸识别 CNN 的抗噪范式 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FHu_Noise-Tolerant_Paradigm_for_Training_Face_Recognition_CNNs_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fhuangyangyu\u002FNoiseFace \"Caffe\")\n- 针对代表性不足数据的人脸识别特征迁移学习 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FYin_Feature_Transfer_Learning_for_Face_Recognition_With_Under-Represented_Data_CVPR_2019_paper.pdf \"CVPR2019\")\n- **Led3D**：一种轻量高效、用于识别低质量 3D 人脸的深度方法 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FMu_Led3D_A_Lightweight_and_Efficient_Deep_Approach_to_Recognizing_Low-Quality_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fmuyouhang\u002FLed3D \"NULL\") [[数据集]](http:\u002F\u002Firip.buaa.edu.cn\u002Flock3dface\u002Findex.html)\n- 用于跨模型人脸识别的 R3 对抗网络 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FChen_R3_Adversarial_Network_for_Cross_Model_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\")\n- **MLT**：人脸识别：一项基于多级分类体系的新综述 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1901.00713 \"arXiv2019\")\n- **CDP**：在海量未标注数据中进行共识驱动传播以实现人脸识别 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.01407 \"ECCV2018\") [[代码]](https:\u002F\u002Fgithub.com\u002FXiaohangZhan\u002Fface_recognition_framework \"PyTorch\") [[项目]](http:\u002F\u002Fmmlab.ie.cuhk.edu.hk\u002Fprojects\u002FCDP\u002F)\n- **GhostVLAD**：用于集合型人脸识别的 GhostVLAD [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.09951 \"ACCV2018\")\n- **DocFace+**：身份证件与自拍匹配 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.05620 \"arXiv2018\") [[代码]](https:\u002F\u002Fgithub.com\u002FseasonSH\u002FDocFace \"TensorFlow\")\n- **2018 综述**：从传统到深度学习的人脸识别 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.00116 \"arXiv2018\")\n- **2018 综述**：深度面部表情识别：一项综述 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.08348 \"arXiv2018\")\n- **2018 综述**：深度人脸识别：一项综述 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.06655 \"arXiv2018\")\n- **SphereFace+(MHE)**：向最小超球能量学习 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1805.09298 \"arXiv2018\") [[代码]](https:\u002F\u002Fgithub.com\u002Fwy1iu\u002Fsphereface-plus \"Caffe\u002FMatlab\")\n- **MobileFace**：移动端的人脸识别解决方案 [[代码]](https:\u002F\u002Fgithub.com\u002FbecauseofAI\u002FMobileFace)\n- **MobileFaceNets**：用于移动设备上准确实时人脸验证的高效 CNN [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1804.07573 \"arXiv2018\") [[代码1]](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface \"MXNet\") [[代码2]](https:\u002F\u002Fgithub.com\u002FKaleidoZhouYN\u002Fmobilefacenet-caffe \"Caffe\") [[代码3]](https:\u002F\u002Fgithub.com\u002Fxsr-ai\u002FMobileFaceNet_TF \"TensorFlow\") [[代码4]](https:\u002F\u002Fgithub.com\u002FGRAYKEY\u002Fmobilefacenet_ncnn \"NCNN\")\n- **FaceID**：使用 RGBD 图像上的人脸嵌入和暹罗网络实现 iPhone X 的 FaceID。[[代码]](https:\u002F\u002Fgithub.com\u002Fnormandipalo\u002FfaceID_beta \"Keras\") [[博客]](https:\u002F\u002Ftowardsdatascience.com\u002Fhow-i-implemented-iphone-xs-faceid-using-deep-learning-in-python-d5dbaa128e1d \"Medium\")\n- **InsightFace(ArcFace)**：2D 和 3D 人脸分析项目 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.07698 \"ArcFace：深度人脸识别的加性角度边际损失(arXiv)\") [[代码1]](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface \"MXNet\")[![GitHub 星标](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fdeepinsight\u002Finsightface.svg?logo=github&label=Stars)](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface) [[代码2]](https:\u002F\u002Fgithub.com\u002Fauroua\u002FInsightFace_TF \"TensorFlow\")\n- **AAM-Softmax(CCL)**：通过集中式坐标学习实现人脸识别 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.05678 \"arXiv2018\")\n- **AM-Softmax**：用于人脸验证的加性边际 Softmax [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.05599 \"arXiv2018\") [[代码1]](https:\u002F\u002Fgithub.com\u002Fhappynear\u002FAMSoftmax \"Caffe\") [[代码2]](https:\u002F\u002Fgithub.com\u002FJoker316701882\u002FAdditive-Margin-Softmax \"TensorFlow\")\n- **CosFace**：用于深度人脸识别的大边际余弦损失 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1801.09414 \"CVPR2018\") [[代码1]](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface \"MXNet\") [[代码2]](https:\u002F\u002Fgithub.com\u002Fyule-li\u002FCosFace \"TensorFlow\")\n- **FeatureIncay**：用于表征正则化的 Feature Incay [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1705.10284 \"ICLR2018\")\n- **CocoLoss**：重新思考大规模识别中的特征判别与聚合 [[论文]](http:\u002F\u002Fcn.arxiv.org\u002Fabs\u002F1710.00870 \"NIPS2017\") [[代码]](https:\u002F\u002Fgithub.com\u002Fsciencefans\u002Fcoco_loss \"Caffe\")\n- **NormFace**：用于人脸验证的 L2 超球嵌入 [[论文]](http:\u002F\u002Fwww.cs.jhu.edu\u002F~alanlab\u002FPubs17\u002Fwang2017normface.pdf \"ACM2017 多媒体会议\") [[代码]](https:\u002F\u002Fgithub.com\u002Fhappynear\u002FNormFace \"Caffe\")\n- **SphereFace(A-Softmax)**：用于人脸识别的深度超球嵌入 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2017\u002Fpapers\u002FLiu_SphereFace_Deep_Hypersphere_CVPR_2017_paper.pdf \"CVPR2017\") [[代码]](https:\u002F\u002Fgithub.com\u002Fwy1iu\u002Fsphereface \"Caffe\")\n- **L-Softmax**：用于卷积神经网络的大边际 Softmax 损失 [[论文]](http:\u002F\u002Fproceedings.mlr.press\u002Fv48\u002Fliud16.pdf \"ICML2016\") [[代码1]](https:\u002F\u002Fgithub.com\u002Fwy1iu\u002FLargeMargin_Softmax_Loss \"Caffe\") [[代码2]](https:\u002F\u002Fgithub.com\u002Fluoyetx\u002Fmx-lsoftmax \"MXNet\") [[代码3]](https:\u002F\u002Fgithub.com\u002FHiKapok\u002Ftf.extra_losses \"TensorFlow\") [[代码4]](https:\u002F\u002Fgithub.com\u002Fauroua\u002FL_Softmax_TensorFlow \"TensorFlow\") [[代码5]](https:\u002F\u002Fgithub.com\u002Ftpys\u002Fface-recognition-caffe2 \"Caffe2\") [[代码6]](https:\u002F\u002Fgithub.com\u002Famirhfarzaneh\u002Flsoftmax-pytorch \"PyTorch\") [[代码7]](https:\u002F\u002Fgithub.com\u002Fjihunchoi\u002Flsoftmax-pytorch \"PyTorch\")\n- **CenterLoss**：一种用于深度人脸识别的判别式特征学习方法 [[论文]](https:\u002F\u002Fydwen.github.io\u002Fpapers\u002FWenECCV16.pdf \"ECCV2016\") [[代码1]](https:\u002F\u002Fgithub.com\u002Fydwen\u002Fcaffe-face \"Caffe\") [[代码2]](https:\u002F\u002Fgithub.com\u002Fpangyupo\u002Fmxnet_center_loss \"MXNet\") [[代码3]](https:\u002F\u002Fgithub.com\u002FShownX\u002Fmxnet-center-loss \"MXNet-Gluon\") [[代码4]](https:\u002F\u002Fgithub.com\u002FEncodeTS\u002FTensorFlow_Center_Loss \"TensorFlow\")\n- **OpenFace**：一个具有移动应用的通用人脸识别库 [[报告]](http:\u002F\u002Felijah.cs.cmu.edu\u002FDOCS\u002FCMU-CS-16-118.pdf \"CMU2016\") [[项目]](http:\u002F\u002Fcmusatyalab.github.io\u002Fopenface\u002F) [[代码1]](https:\u002F\u002Fgithub.com\u002Fcmusatyalab\u002Fopenface \"Torch\") [[代码2]](https:\u002F\u002Fgithub.com\u002Fthnkim\u002FOpenFacePytorch \"PyTorch\")\n- **FaceNet**：用于人脸识别和聚类的统一嵌入 [[论文]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fpapers\u002FSchroff_FaceNet_A_Unified_2015_CVPR_paper.pdf \"CVPR2015\") [[代码]](https:\u002F\u002Fgithub.com\u002Fdavidsandberg\u002Ffacenet \"TensorFlow\")\n- **DeepID3**：使用非常深的神经网络进行人脸识别 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1502.00873 \"arXiv2015\")\n- **DeepID2+**：深度学习得到的人脸表征是稀疏、选择性和鲁棒的 [[论文]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2015\u002Fpapers\u002FSun_Deeply_Learned_Face_2015_CVPR_paper.pdf \"CVPR2015\")\n- **DeepID2**：通过联合身份识别与验证实现深度学习的人脸表征 [[论文]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F5416-deep-learning-face-representation-by-joint-identification-verification.pdf \"NIPS2014\")\n- **DeepID**：通过预测 10,000 个类别实现深度学习的人脸表征 [[论文]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fpapers\u002FSun_Deep_Learning_Face_2014_CVPR_paper.pdf \"CVPR2014\")\n- **DeepFace**：在人脸验证方面缩小与人类水平性能的差距 [[论文]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fpapers\u002FTaigman_DeepFace_Closing_the_2014_CVPR_paper.pdf \"CVPR2014\")\n- **LBP+Joint Bayes**：重访贝叶斯人脸：一种联合公式 [[论文]](https:\u002F\u002Fs3.amazonaws.com\u002Facademia.edu.documents\u002F31414608\u002FJointBayesian.pdf?AWSAccessKeyId=AKIAIWOWYYGZ2Y53UL3A&Expires=1543656042&Signature=k6LefuQnIC2x8gep7yQTxqKgzus%3D&response-content-disposition=inline%3B%20filename%3DBayesian_Face_Revisited_A_Joint_Formulat.pdf \"ECCV2012\") [[代码1]](https:\u002F\u002Fgithub.com\u002Fcyh24\u002FJoint-Bayesian \"Python\") [[代码2]](https:\u002F\u002Fgithub.com\u002FMaoXu\u002FJoint_Bayesian \"Matlab\") [[代码3]](https:\u002F\u002Fgithub.com\u002FGlasssix\u002Fjoint_bayesian \"C++\u002FC#\")\n- **LBPFace**：使用局部二值模式进行人脸识别 [[论文]](https:\u002F\u002Fpdfs.semanticscholar.org\u002F3242\u002F0c65f8ef0c5bd83b14c8ae662cbce73e6781.pdf \"ECCV2004\") [[代码]](https:\u002F\u002Fdocs.opencv.org\u002F2.4\u002Fmodules\u002Fcontrib\u002Fdoc\u002Ffacerec\u002Ffacerec_tutorial.html \"OpenCV\")\n- **FisherFace(LDA)**：特征脸 vs. Fisher脸：使用类特定线性投影进行识别 [[论文]](https:\u002F\u002Fapps.dtic.mil\u002Fdtic\u002Ftr\u002Ffulltext\u002Fu2\u002F1015508.pdf \"TPAMI1997\") [[代码]](https:\u002F\u002Fdocs.opencv.org\u002F2.4\u002Fmodules\u002Fcontrib\u002Fdoc\u002Ffacerec\u002Ffacerec_tutorial.html \"OpenCV\")\n- **EigenFace(PCA)**：使用特征脸进行人脸识别 [[论文]](http:\u002F\u002Fwww.cs.ucsb.edu\u002F~mturk\u002FPapers\u002Fmturk-CVPR91.pdf \"CVPR1991\") [[代码]](https:\u002F\u002Fdocs.opencv.org\u002F2.4\u002Fmodules\u002Fcontrib\u002Fdoc\u002Ffacerec\u002Ffacerec_tutorial.html \"OpenCV\")\n\n## 🔖 人脸检测\n- **RetinaFace**: 野外单阶段密集人脸定位 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.00641 \"arXiv2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fdeepinsight\u002Finsightface\u002Ftree\u002Fmaster\u002FRetinaFace \"MXNet\")\n- 尺度不变的人脸检测中的分组采样 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FMing_Group_Sampling_for_Scale_Invariant_Face_Detection_CVPR_2019_paper.pdf \"CVPR2019\")\n- **FA-RPN**: 用于人脸检测的浮动区域建议框 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FNajibi_FA-RPN_Floating_Region_Proposals_for_Face_Detection_CVPR_2019_paper.pdf \"CVPR2019\")\n- **SFA**: 小人脸注意力人脸检测器 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1812.08402 \"SPIC2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fshiluo1990\u002FSFA \"Caffe\")\n- **ISRN**: 改进的选择性精炼网络用于人脸检测 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1901.06651 \"arXiv2019\")\n- **DSFD**: 双重射击人脸检测器 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1810.10220 \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FTencentYoutuResearch\u002FFaceDetection-DSFD \"PyTorch\")\n- **PyramidBox++**: 高性能微小人脸检测器 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.00386 \"arXiv2019\")\n- **VIM-FD**: 鲁棒且高性能的人脸检测器 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1901.02350 \"arXiv2019\")\n- **SHF**: 通过在困难图像上学习小人脸实现鲁棒的人脸检测 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1811.11662 \"arXiv2018\") [[代码]](https:\u002F\u002Fgithub.com\u002Fbairdzhang\u002Fsmallhardface \"Caffe\")\n- **SRN**: 用于高性能人脸检测的选择性精炼网络 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1809.02693 \"AAAI2019\")\n- **SFDet**: 单次尺度感知网络用于实时人脸检测 [[论文]](https:\u002F\u002Flink.springer.com\u002Fepdf\u002F10.1007\u002Fs11263-019-01159-3?author_access_token=Jjgl-u1CAXPmSKWDljfSBfe4RwlQNchNByi7wbcMAY7Vwo_nrkuFMElF6YSQ0We34tUs42D0dyurcBAD0sJP66n6GBanVgA9qsuvh4Y_Bjf3E_n9_croQ4esS882srfHyUz-L96pU3gu_M30Kk6_XQ%3D%3D \"IJCV2019\")\n- **HyperFace**: 一个用于人脸检测、关键点定位、姿态估计和性别识别的深度多任务学习框架 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1603.01249 \"TPAMI2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fmaharshi95\u002FHyperFace \"TensorFlow\")\n- **PyramidBox**: 一种上下文辅助的单次人脸检测器 [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1803.07737.pdf \"arXiv2018\") [[代码]](https:\u002F\u002Fgithub.com\u002FPaddlePaddle\u002Fmodels\u002Ftree\u002F2a6b7dc92f04815f0b298e59030cb779dd0e038c\u002Ffluid\u002Fface_detction \"PaddlePaddle\")\n- **PCN**: 基于渐进校准网络的实时旋转不变人脸检测 [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1804.06039.pdf \"CVPR2018\") [[代码]](https:\u002F\u002Fgithub.com\u002FJack-CV\u002FPCN \"C++\")\n- **S³FD**: 单次尺度不变人脸检测器 [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1708.05237.pdf \"arXiv2017\") [[代码]](https:\u002F\u002Fgithub.com\u002Fsfzhang15\u002FSFD \"Caffe\")\n- **SSH**: 单阶段无头人脸检测器 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2017\u002Fpapers\u002FNajibi_SSH_Single_Stage_ICCV_2017_paper.pdf \"ICCV2017\") [[代码]](https:\u002F\u002Fgithub.com\u002Fmahyarnajibi\u002FSSH \"Caffe\")\n- **FaceBoxes**: 一款高精度的CPU实时人脸检测器 [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1708.05234.pdf \"IJCB2017\")[[代码1]](https:\u002F\u002Fgithub.com\u002Fzeusees\u002FFaceBoxes \"Caffe\") [[代码2]](https:\u002F\u002Fgithub.com\u002Flxg2015\u002Ffaceboxes \"PyTorch\")\n- **TinyFace**: 检测微小人脸 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2017\u002Fpapers\u002FHu_Finding_Tiny_Faces_CVPR_2017_paper.pdf \"CVPR2017\") [[项目]](https:\u002F\u002Fwww.cs.cmu.edu\u002F~peiyunh\u002Ftiny\u002F) [[代码1]](https:\u002F\u002Fgithub.com\u002Fpeiyunh\u002Ftiny \"MatConvNet\") [[代码2]](https:\u002F\u002Fgithub.com\u002Fchinakook\u002Fhr101_mxnet \"MXNet\") [[代码3]](https:\u002F\u002Fgithub.com\u002Fcydonia999\u002FTiny_Faces_in_Tensorflow \"TensorFlow\")\n- **MTCNN**: 使用多任务级联卷积网络进行联合人脸检测与对齐 [[论文]](https:\u002F\u002Fkpzhang93.github.io\u002FMTCNN_face_detection_alignment\u002Fpaper\u002Fspl.pdf \"SPL2016\") [[项目]](https:\u002F\u002Fkpzhang93.github.io\u002FMTCNN_face_detection_alignment\u002F) [[代码1]](https:\u002F\u002Fgithub.com\u002Fkpzhang93\u002FMTCNN_face_detection_alignment \"Caffe\") [[代码2]](https:\u002F\u002Fgithub.com\u002FCongWeilin\u002Fmtcnn-caffe \"Caffe\") [[代码3]](https:\u002F\u002Fgithub.com\u002FforeverYoungGitHub\u002FMTCNN \"Caffe\") [[代码4]](https:\u002F\u002Fgithub.com\u002FSeanlinx\u002Fmtcnn \"MXNet\") [[代码5]](https:\u002F\u002Fgithub.com\u002Fpangyupo\u002Fmxnet_mtcnn_face_detection \"MXNet\") [[代码6]](https:\u002F\u002Fgithub.com\u002FTropComplique\u002Fmtcnn-pytorch \"PyTorch\") [[代码7]](https:\u002F\u002Fgithub.com\u002FAITTSMD\u002FMTCNN-Tensorflow \"TensorFlow\")\n- **NPD**: 一款快速且准确的非约束条件下的人脸检测器 [[论文]](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fusers\u002Fscliao\u002Fpapers\u002FLiao-PAMI15-NPD.pdf \"TPAMI2015\") [[代码]](https:\u002F\u002Fgithub.com\u002Fwincle\u002FNPD \"C++\") [[项目]](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fusers\u002Fscliao\u002Fprojects\u002Fnpdface\u002Findex.html)\n- **PICO**: 基于决策树组织的像素强度比较目标检测 [[论文]](https:\u002F\u002Farxiv.org\u002Fpdf\u002F1305.4537.pdf \"arXiv2014\") [[代码]](https:\u002F\u002Fgithub.com\u002Fnenadmarkus\u002Fpico \"C\")\n- **libfacedetection**: 一个用于图像中人脸检测和人脸关键点检测的快速二进制库。[[代码]](https:\u002F\u002Fgithub.com\u002FShiqiYu\u002Flibfacedetection \"C++\")\n- **SeetaFaceEngine**: SeetaFace检测、SeetaFace对齐和SeetaFace识别 [[代码]](https:\u002F\u002Fgithub.com\u002Fseetaface\u002FSeetaFaceEngine \"C++\")\n\n## 🔖 人脸关键点定位\n- **DeCaFA**: 用于野外人脸对齐的深度卷积级联网络 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.02549)\n- **AWing**: 基于热图回归的鲁棒人脸对齐自适应翼损失函数 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.07399 \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fprotossw512\u002FAdaptiveWingLoss \"PyTorch\")\n- **KDN**: 基于核密度估计的深度神经网络人脸对齐 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FChen_Face_Alignment_With_Kernel_Density_Deep_Neural_Network_ICCV_2019_paper.pdf \"ICCV2019\")\n- **语义对齐**: 寻找面部关键点检测的语义一致真值 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FLiu_Semantic_Alignment_Finding_Semantically_Consistent_Ground-Truth_for_Facial_Landmark_Detection_CVPR_2019_paper.pdf \"CVPR2019\")\n- 基于遮挡自适应深度网络的鲁棒面部关键点检测 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhu_Robust_Facial_Landmark_Detection_via_Occlusion-Adaptive_Deep_Networks_CVPR_2019_paper.pdf \"CVPR2019\")\n- **PFLD**: 一种实用的面部关键点检测器 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.10859 \"arXiv2019\") [[项目]](https:\u002F\u002Fsites.google.com\u002Fview\u002Fxjguo\u002Ffld) [[代码]](https:\u002F\u002Fdrive.google.com\u002Ffile\u002Fd\u002F1n1uZPbM9Wz052aVnlc_3L4gjQHiwfj4B\u002Fview \"APK\")\n- **PRNet**: 结合位置图回归网络的三维人脸重建与密集对齐 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ECCV_2018\u002Fpapers\u002FYao_Feng_Joint_3D_Face_ECCV_2018_paper.pdf \"ECCV2018\") [[代码]](https:\u002F\u002Fgithub.com\u002FYadiraF\u002FPRNet \"TensorFlow\")\n- **LAB**: 关注边界：一种基于边界的面部对齐算法 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FWu_Look_at_Boundary_CVPR_2018_paper.pdf \"CVPR2018\") [[项目]](https:\u002F\u002Fwywu.github.io\u002Fprojects\u002FLAB\u002FLAB.html) [[代码]](https:\u002F\u002Fgithub.com\u002Fwywu\u002FLAB \"Caffe\")\n- **Face-Alignment**: 我们距离解决二维和三维人脸对齐问题还有多远？（附带包含23万个三维人脸关键点的数据集）[[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2017\u002Fpapers\u002FBulat_How_Far_Are_ICCV_2017_paper.pdf \"ICCV2017\") [[项目]](https:\u002F\u002Fadrianbulat.com\u002Fface-alignment) [[代码1]](https:\u002F\u002Fgithub.com\u002F1adrianb\u002Fface-alignment \"PyTorch\") [[代码2]](https:\u002F\u002Fgithub.com\u002F1adrianb\u002F2D-and-3D-face-alignment \"Torch7\")\n- **ERT**: 基于回归树集成的一毫秒人脸对齐 [[论文]](https:\u002F\u002Fwww.cv-foundation.org\u002Fopenaccess\u002Fcontent_cvpr_2014\u002Fpapers\u002FKazemi_One_Millisecond_Face_2014_CVPR_paper.pdf \"CVPR2014\") [[代码]](http:\u002F\u002Fdlib.net\u002Fimaging.html \"Dlib\")\n\n## 🔖 人脸聚类\n- **BCL**: 面向聚类数目未知的视频人脸聚类 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.03381 \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fmakarandtapaswi\u002FBallClustering_ICCV2019 \"PyTorch\")\n- **LinkageFace**: 基于图卷积网络的链接法人脸聚类 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.11306 \"CVPR2019\")\n- **LTC**: 在亲和图上学习进行人脸聚类 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FYang_Learning_to_Cluster_Faces_on_an_Affinity_Graph_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fyl-1993\u002Flearn-to-cluster \"PyTorch\")\n\n## 🔖 人脸表情\n- **FECNet**: 用于面部表情相似度计算的紧凑嵌入 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FVemulapalli_A_Compact_Embedding_for_Facial_Expression_Similarity_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FGerardLiu96\u002FFECNet \"Keras\")\n- **LBVCNN**: 用于从图像序列中识别面部表情的局部二值体积卷积神经网络 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.07647 \"arXiv2019\")\n\n## 🔖 人脸动作单元\n- 用于面部动作单元强度估计的联合表示与估计器学习 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhang_Joint_Representation_and_Estimator_Learning_for_Facial_Action_Unit_Intensity_CVPR_2019_paper.pdf \"CVPR2019\")\n- 基于特定人形状正则化的局部关系学习用于面部动作单元检测 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FNiu_Local_Relationship_Learning_With_Person-Specific_Shape_Regularization_for_Facial_Action_CVPR_2019_paper.pdf \"CVPR2019\")\n- **TCAE**: 基于视频的自监督表示学习用于面部动作单元检测 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FLi_Self-Supervised_Representation_Learning_From_Videos_for_Facial_Action_Unit_Detection_CVPR_2019_paper.pdf \"CVPR2019 口头报告\") [[代码]](https:\u002F\u002Fgithub.com\u002Fmysee1989\u002FTCAE \"PyTorch\")\n- **JAANet**: 用于联合面部动作单元检测与人脸对齐的深度自适应注意力机制 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ECCV_2018\u002Fpapers\u002FZhiwen_Shao_Deep_Adaptive_Attention_ECCV_2018_paper.pdf \"ECCV2018\") [[代码]](https:\u002F\u002Fgithub.com\u002FZhiwenShao\u002FJAANet \"Caffe\")\n\n## 🔖 人脸3D\n- **DF2Net**: 一种用于精细3D人脸重建的密集-精细-更精细网络 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FZeng_DF2Net_A_Dense-Fine-Finer_Network_for_Detailed_3D_Face_Reconstruction_ICCV_2019_paper.pdf \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fxiaoxingzeng\u002FDF2Net \"PyTorch\")\n- 基于端到端形状保持域迁移的半监督单目3D人脸重建 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FPiao_Semi-Supervised_Monocular_3D_Face_Reconstruction_With_End-to-End_Shape-Preserved_Domain_Transfer_ICCV_2019_paper.pdf \"ICCV2019\")\n- **3DFC**: 从多样化原始扫描数据中进行3D人脸建模 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.04943 \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fliuf1990\u002F3DFC \"PyTorch\")\n- **2DASL**: 基于2D辅助自监督学习，从单张图像联合进行3D人脸重建和密集人脸对齐 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1903.09359 \"arXiv2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FXgTu\u002F2DASL \"PyTorch & Matlab\")\n- **MVF-Net**: 多视角3D人脸可变形模型回归 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FWu_MVF-Net_Multi-View_3D_Face_Morphable_Model_Regression_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FFanziapril\u002Fmvfnet \"PyTorch\")\n- 高达2500FPS的密集3D人脸解码：联合纹理与形状的卷积网格解码器 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhou_Dense_3D_Face_Decoding_Over_2500FPS_Joint_Texture__Shape_CVPR_2019_paper.pdf \"CVPR2019\")\n- 向高保真非线性3D人脸可变形模型迈进 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FTran_Towards_High-Fidelity_Nonlinear_3D_Face_Morphable_Model_CVPR_2019_paper.pdf \"CVPR2019\") [[项目]](http:\u002F\u002Fcvlab.cse.msu.edu\u002Fproject-nonlinear-3dmm.html)\n- 组合3D可变形模型：大规模面部与头部模型 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FPloumpis_Combining_3D_Morphable_Models_A_Large_Scale_Face-And-Head_Model_CVPR_2019_paper.pdf \"CVPR2019\")\n- 3D人脸形状的解耦表征学习 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FJiang_Disentangled_Representation_Learning_for_3D_Face_Shape_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FzihangJiang\u002FDR-Learning-for-3D-Face \"Keras\")\n- 高保真人脸模型的自监督适应，用于单目表演跟踪 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FYoon_Self-Supervised_Adaptation_of_High-Fidelity_Face_Models_for_Monocular_Performance_Tracking_CVPR_2019_paper.pdf \"CVPR2019\")\n- **MMFace**: 一种用于无约束人脸重建的多指标回归网络 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FYi_MMFace_A_Multi-Metric_Regression_Network_for_Unconstrained_Face_Reconstruction_CVPR_2019_paper.pdf \"CVPR2019\")\n- **RingNet**: 学习从图像中回归3D人脸形状和表情，无需3D监督 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FSanyal_Learning_to_Regress_3D_Face_Shape_and_Expression_From_an_Image_without_3D_Supervision_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fsoubhiksanyal\u002FRingNet \"TensorFlow\") [[项目]](https:\u002F\u002Fringnet.is.tue.mpg.de\u002F)\n- 提升局部形状匹配以实现密集3D人脸对应 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FFan_Boosting_Local_Shape_Matching_for_Dense_3D_Face_Correspondence_CVPR_2019_paper.pdf \"CVPR2019\")\n- **FML**: 从视频中学习人脸模型 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FTewari_FML_Face_Model_Learning_From_Videos_CVPR_2019_paper.pdf \"CVPR2019\")\n\n## 🔖 人脸生成对抗网络\n#### 人脸特征\n- 面部到参数的转换用于游戏角色自动生成 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1909.01064 \"ICCV2019\")\n---\n#### 人脸编辑\n- **SC-FEGAN**: 基于用户草图和颜色的人脸编辑生成对抗网络 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1902.06838 \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Frun-youngjoo\u002FSC-FEGAN \"TensorFlow\")\n---\n#### 人脸去遮挡\n- 利用3D可变形模型和生成对抗网络进行人脸去遮挡 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.06109 \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fxweiyuan\u002FFace-de-occlusion-using-3D-morphable-model-and-generative-adversarial-network)\n---\n#### 人脸年龄变化  \n- 通过深度强化学习实现视频中的人脸自动老化 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FDuong_Automatic_Face_Aging_in_Videos_via_Deep_Reinforcement_Learning_CVPR_2019_paper.pdf \"CVPR2019\") [[博客]](https:\u002F\u002Fwww.fastcompany.com\u002F90314606\u002Fthis-new-ai-tool-makes-creepily-realistic-videos-of-faces-in-the-future)\n- 基于小波的生成对抗网络实现属性感知的人脸老化 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FLiu_Attribute-Aware_Face_Aging_With_Wavelet-Based_Generative_Adversarial_Networks_CVPR_2019_paper.pdf \"CVPR2019\")\n- **SAGAN**: 用于人脸属性编辑的空间注意力生成对抗网络 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ECCV_2018\u002Fpapers\u002FGang_Zhang_Generative_Adversarial_Network_ECCV_2018_paper.pdf \"ECCV2018\") [[代码]](https:\u002F\u002Fgithub.com\u002Felvisyjlin\u002FSpatialAttentionGAN \"PyTorch\")\n---\n#### 人脸绘画 \n- **APDrawingGAN**: 基于分层GAN从人脸照片生成艺术肖像画 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FYi_APDrawingGAN_Generating_Artistic_Portrait_Drawings_From_Face_Photos_With_Hierarchical_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fyiranran\u002FAPDrawingGAN \"PyTorch\")\n---\n#### 人脸生成\n- **StyleGAN**: 基于风格的生成器架构用于生成对抗网络 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FKarras_A_Style-Based_Generator_Architecture_for_Generative_Adversarial_Networks_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fstylegan \"TensorFlow\") [[数据集]](https:\u002F\u002Fgithub.com\u002FNVlabs\u002Fffhq-dataset \"FFHQ\")\n---\n#### 人脸化妆\n- **BeautyGAN**: 基于深度生成对抗网络的实例级面部化妆迁移 [[论文]](http:\u002F\u002Fliusi-group.com\u002Fpdf\u002FBeautyGAN-camera-ready_2.pdf \"多媒体会议, ACM2018\") [[代码]](https:\u002F\u002Fgithub.com\u002FHonlan\u002FBeautyGAN \"TensorFlow\") [[项目]](http:\u002F\u002Fliusi-group.com\u002Fprojects\u002FBeautyGAN) [[海报]](http:\u002F\u002Fliusi-group.com\u002Fpdf\u002FBeautyGAN-camera-ready_2_poster.pdf)\n---\n#### 人脸交换\n- **FSGAN**: 主体无关的人脸交换与重演 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.05932 \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FYuvalNirkin) [[项目]](https:\u002F\u002Fnirkin.com\u002Ffsgan\u002F)\n- **Faceswap**: 一种利用深度学习识别并交换图片和视频中人脸的工具 [[代码1]](https:\u002F\u002Fgithub.com\u002Fdeepfakes\u002Ffaceswap \"TensorFlow\") [[代码2]](https:\u002F\u002Fgithub.com\u002Fiperov\u002FDeepFaceLab \"TensorFlow\u002FKeras\")\n- **FUNIT**: 少样本无监督图像到图像转换 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1905.01723 \"arXiv2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FNVlabs\u002FFUNIT \"PyTorch\") [[项目]](https:\u002F\u002Fnvlabs.github.io\u002FFUNIT\u002F)\n---\n#### 其他人脸相关\n- 野外极端姿态和表情下的无监督人脸归一化 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FQian_Unsupervised_Face_Normalization_With_Extreme_Pose_and_Expression_in_the_Wild_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fmx54039q\u002Ffnm \"TensorFlow\")\n- **GANFIT**: 用于高保真3D人脸重建的生成对抗网络拟合 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FGecer_GANFIT_Generative_Adversarial_Network_Fitting_for_High_Fidelity_3D_Face_CVPR_2019_paper.pdf \"CVPR2019\") [[项目]](https:\u002F\u002Fgithub.com\u002Fbarisgecer\u002FGANFit)\n- **HF-PIM**: 学习高分辨率人脸正面化的高保真姿态不变模型 [[论文]](http:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F7551-learning-a-high-fidelity-pose-invariant-model-for-high-resolution-face-frontalization.pdf \"NIPS2018\")\n- **Super-FAN**: 结合GAN对任意姿态的真实世界低分辨率人脸进行联合地标定位与超分辨率 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FBulat_Super-FAN_Integrated_Facial_CVPR_2018_paper.pdf \"CVPR2018 Spotlight\")\n- **GANimation**: 基于单张图像的解剖学感知面部动画 [[论文]](https:\u002F\u002Fwww.albertpumarola.com\u002Fpublications\u002Ffiles\u002Fpumarola2018ganimation.pdf \"ECCV2018 口头报告，最佳论文奖荣誉提及\") [[项目]](https:\u002F\u002Fwww.albertpumarola.com\u002Fresearch\u002FGANimation\u002Findex.html) [[代码]](https:\u002F\u002Fgithub.com\u002Falbertpumarola\u002FGANimation \"PyTorch\")\n- **StarGAN**: 用于多领域图像到图像转换的统一生成对抗网络 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_cvpr_2018\u002Fpapers\u002FChoi_StarGAN_Unified_Generative_CVPR_2018_paper.pdf \"CVPR2018\")\n[[代码]](https:\u002F\u002Fgithub.com\u002Fyunjey\u002FStarGAN \"PyTorch\")\n- **PGAN**: 用于提升质量、稳定性和多样性的渐进式GAN增长 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1710.10196 \"ICLR2018\")\n[[代码1]](https:\u002F\u002Fgithub.com\u002Ftkarras\u002Fprogressive_growing_of_gans \"TensorFlow\") [[代码2]](https:\u002F\u002Fgithub.com\u002Fgithub-pengge\u002FPyTorch-progressive_growing_of_gans \"PyTorch\")\n\n## 🔖 人脸去模糊\n- 基于3D面部先验的人脸视频去模糊 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_ICCV_2019\u002Fpapers\u002FRen_Face_Video_Deblurring_Using_3D_Facial_Priors_ICCV_2019_paper.pdf \"ICCV2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Frwenqi\u002F3Dfacedeblurring \"TensorFlow\")\n\n## 🔖 人脸超分辨率\n- **PFSR**: 基于面部地标注意力的渐进式人脸超分辨率 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.08239 \"BMVC2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FDeokyunKim\u002FProgressive-Face-Super-Resolution \"PyTorch\")\n\n## 🔖 人脸操控\n- **Make a Face**: 向任意高保真度的人脸操控迈进 [[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1908.07191 \"ICCV2019\")\n- 3D引导的细粒度人脸操控 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FGeng_3D_Guided_Fine-Grained_Face_Manipulation_CVPR_2019_paper.pdf \"CVPR2019\")\n- **SemanticComponent**: 用于人脸属性操控的语义组件分解 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FChen_Semantic_Component_Decomposition_for_Face_Attribute_Manipulation_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fyingcong\u002FSemanticComponent) [[演示]](http:\u002F\u002Fappsrv.cse.cuhk.edu.hk\u002F~ycchen\u002Fdemos\u002Fsemantic_component.mp4)\n\n## 🔖 人脸防伪\n- **数据集与基准测试**：大规模多模态人脸防伪数据集与基准测试 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FZhang_A_Dataset_and_Benchmark_for_Large-Scale_Multi-Modal_Face_Anti-Spoofing_CVPR_2019_paper.pdf \"CVPR2019\") [[海报]](http:\u002F\u002Fwww.cbsr.ia.ac.cn\u002Fusers\u002Fsfzhang\u002FShifeng%20Zhang's%20Homepage_files\u002FCVPR2019_CASIA-SURF_Poster.pdf) [[数据集]](https:\u002F\u002Fsites.google.com\u002Fqq.com\u002Fchalearnfacespoofingattackdete\u002F)\n- 面向零样本学习的人脸防伪深度树学习 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FLiu_Deep_Tree_Learning_for_Zero-Shot_Face_Anti-Spoofing_CVPR_2019_paper.pdf \"CVPR2019 口头报告\") \n\n## 🔖 人脸对抗攻击\n- 用于年龄不变人脸识别的去相关对抗学习 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FWang_Decorrelated_Adversarial_Learning_for_Age-Invariant_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\") \n- 用于人脸呈现攻击检测的多对抗判别式深度领域泛化 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FShao_Multi-Adversarial_Discriminative_Deep_Domain_Generalization_for_Face_Presentation_Attack_Detection_CVPR_2019_paper.pdf \"CVPR2019\") \n- 面向人脸识别的高效决策型黑盒对抗攻击 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FDong_Efficient_Decision-Based_Black-Box_Adversarial_Attacks_on_Face_Recognition_CVPR_2019_paper.pdf \"CVPR2019\") \n\n## 🔖 跨模态人脸\n- **从推理到生成**：端到端完全自监督的语音驱动人脸生成 [[论文]](https:\u002F\u002Fopenreview.net\u002Fpdf?id=H1guaREYPr \"ICLR2020\")\n- **FRV**：基于生成对抗网络的语音驱动人脸重建 [[论文]](https:\u002F\u002Fpapers.nips.cc\u002Fpaper\u002F8768-face-reconstruction-from-voice-using-generative-adversarial-networks.pdf \"NeurIPS2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fcmu-mlsp\u002Freconstructing_faces_from_voices \"PyTorch\") [[海报]](https:\u002F\u002Fydwen.github.io\u002Fpapers\u002FWenNeurIPS19-poster.pdf)\n- **Speech2Face**：学习声音背后的脸 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FOh_Speech2Face_Learning_the_Face_Behind_a_Voice_CVPR_2019_paper.pdf \"CVPR2019\") [[项目]](https:\u002F\u002Fspeech2face.github.io\u002F)\n- **JFDFMR**：多人脸联合检测与面部动作重定向 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FChaudhuri_Joint_Face_Detection_and_Facial_Motion_Retargeting_for_Multiple_Faces_CVPR_2019_paper.pdf \"CVPR2019\")\n- **ATVGnet**：具有动态像素级损失的分层跨模态说话人脸生成 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FChen_Hierarchical_Cross-Modal_Talking_Face_Generation_With_Dynamic_Pixel-Wise_Loss_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Flelechen63\u002FATVGnet \"PyTorch\")\n\n## 🔖 人脸捕捉\n- 基于解剖肌肉的高质量人脸捕捉 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FBao_High-Quality_Face_Capture_Using_Anatomical_Muscles_CVPR_2019_paper.pdf \"CVPR2019\")\n- 单目全身捕捉：在自然场景中捕捉面部、身体和双手的姿态 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FXiang_Monocular_Total_Capture_Posing_Face_Body_and_Hands_in_the_Wild_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002FCMU-Perceptual-Computing-Lab\u002FMonocularTotalCapture) [[项目]](http:\u002F\u002Fdomedb.perception.cs.cmu.edu\u002Fmtc.html)\n- 富有表现力的身体捕捉：从单张图像重建3D手部、面部和身体 [[论文]](http:\u002F\u002Fopenaccess.thecvf.com\u002Fcontent_CVPR_2019\u002Fpapers\u002FPavlakos_Expressive_Body_Capture_3D_Hands_Face_and_Body_From_a_Single_Image_CVPR_2019_paper.pdf \"CVPR2019\") [[代码]](https:\u002F\u002Fgithub.com\u002Fvchoutas\u002Fsmplify-x \"PyTorch\") [[项目]](https:\u002F\u002Fsmpl-x.is.tue.mpg.de\u002F)\n\n## :hammer: 人脸库与工具\n- **Dlib** [[网址]](http:\u002F\u002Fdlib.net\u002Fimaging.html \"图像处理\") [[GitHub]](https:\u002F\u002Fgithub.com\u002Fdavisking\u002Fdlib \"master\")\n- **OpenCV** [[文档]](https:\u002F\u002Fdocs.opencv.org \"所有版本\") [[GitHub]](https:\u002F\u002Fgithub.com\u002Fopencv\u002Fopencv\u002F \"master\")\n- **Face3D** [[GitHub]](https:\u002F\u002Fgithub.com\u002FYadiraF\u002Fface3d \"master\")\n\n---\n# \u003Cp align=\"center\">:boom:**大爆炸**:boom:\u003C\u002Fp>\n\n#### \u003Cp align=\"center\">**感受野是天然锚点**\u003C\u002Fp>\n#### \u003Cp align=\"center\">**感受野就是你所需要的全部**\u003C\u002Fp>\n\u003Cp align=\"center\">2K分辨率下的实时检测原来如此简单！\u003C\u002Fp>\n\n\u003Cdiv align=\"center\">\u003Cimg width=\"1280\" height=\"auto\" src=\"https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FbecauseofAI_awesome-face_readme_d8b4e880ba80.gif\"\u002F>\u003C\u002Fdiv>  \n\n#### \u003Cp align=\"center\">[[论文]](https:\u002F\u002Farxiv.org\u002Fabs\u002F1904.10633) [[MXNet]](https:\u002F\u002Fgithub.com\u002FYonghaoHe\u002FA-Light-and-Fast-Face-Detector-for-Edge-Devices) [[PyTorch]](https:\u002F\u002Fgithub.com\u002FbecauseofAI\u002Flffd-pytorch) [![GitHub 星标](https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002FYonghaoHe\u002FA-Light-and-Fast-Face-Detector-for-Edge-Devices.svg?logo=github&label=Stars)](https:\u002F\u002Fgithub.com\u002FYonghaoHe\u002FA-Light-and-Fast-Face-Detector-for-Edge-Devices)\u003C\u002Fp>","# awesome-face (HelloFace) 快速上手指南\n\n`awesome-face` (HelloFace) 是一个全面的人脸技术资源仓库，汇集了人脸检测、识别、对齐、聚类、3D 重建、GAN 生成及对抗攻击等领域的 SOTA（最先进）论文、数据集和开源工具。本项目主要作为**索引和导航**，帮助开发者快速定位所需的技术方案和代码库。\n\n## 环境准备\n\n由于本项目是资源列表而非单一可执行软件，使用前需根据你选择的具体子项目（如人脸检测或识别模型）配置环境。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 3.6+\n*   **深度学习框架**: PyTorch 1.7+ 或 TensorFlow 2.x (视具体子项目而定)\n*   **硬件要求**: 建议配备 NVIDIA GPU (CUDA 10.2+) 以加速模型训练和推理\n*   **前置依赖**:\n    ```bash\n    pip install opencv-python numpy matplotlib scipy\n    ```\n\n> **提示**: 具体的版本依赖请参考仓库中对应子项目（如 `Face Recognition` 或 `Face Detection` 章节下链接的原始论文代码库）的 `requirements.txt`。\n\n## 安装步骤\n\n`awesome-face` 本身无需安装，只需克隆仓库即可获取最新的资源列表和论文索引。\n\n1.  **克隆仓库**\n    ```bash\n    git clone https:\u002F\u002Fgithub.com\u002FbecauseofAI\u002FHelloFace.git\n    cd HelloFace\n    ```\n\n2.  **获取特定模型代码**\n    浏览本仓库的 `README.md` 或访问在线网站 [https:\u002F\u002FbecauseofAI.github.io\u002FHelloFace](https:\u002F\u002FbecauseofAI.github.io\u002FHelloFace)，找到你需要的技术方向（例如 **Face Recognition** 下的 **MagFace**）。\n\n    点击对应论文的 `[paper]` 或代码链接跳转至原始项目页面进行安装。以 **MagFace** 为例：\n    ```bash\n    # 示例：跳转到 MagFace 官方仓库并克隆\n    git clone https:\u002F\u002Fgithub.com\u002FIrvingMeng\u002FMagFace.git\n    cd MagFace\n    pip install -r requirements.txt\n    ```\n\n## 基本使用\n\n本仓库的核心用法是**检索**。以下演示如何利用该仓库快速启动一个人脸识别任务。\n\n### 1. 检索模型\n在 `HelloFace` 的目录结构中，进入 `Face Recognition` 部分，找到 CVPR 2021 的 **MagFace** 项目。\n\n### 2. 运行示例 (以 MagFace 为例)\n假设你已按照上述步骤安装了具体的模型代码，通常的使用流程如下：\n\n**步骤 A: 数据准备**\n```bash\n# 创建数据集目录\nmkdir -p datasets\u002Fms1m-retinaface-tiny\n# 下载数据集 (请参考具体项目的下载链接)\n```\n\n**步骤 B: 模型推理**\n大多数项目提供简单的测试脚本。以下为典型的推理命令结构：\n```bash\npython test.py --backbone resnet50 --network backbone.pth --dataset lfw\n```\n\n**步骤 C: 查看结果**\n程序将输出准确率指标或在 `output\u002F` 目录下生成可视化结果。\n\n### 3. 探索更多资源\n你可以直接在本地 `README.md` 中搜索关键词（如 `Face Swap`, `3D`, `Anti-Spoofing`），快速定位到相关领域的最新论文和 GitHub 链接，直接复用社区的最佳实践。\n\n---\n*注：本指南仅针对 `awesome-face` 资源库的导航使用。具体算法的训练参数、数据预处理细节请严格参照各子项目官方文档。*","某安防科技公司的算法团队正在为新一代智慧园区门禁系统研发高精度人脸识别模块，需处理复杂光照下的人脸检测、关键点定位及身份验证任务。\n\n### 没有 awesome-face 时\n- **技术选型困难**：面对海量且分散的论文与代码库，团队耗费数周时间筛选适合“宽脸（WIDER FACE）”等严苛场景的检测模型，效率极低。\n- **性能基准缺失**：缺乏统一的 SOTA（最先进）指标参考，难以判断自研模型在人脸对齐和验证任务上是否已达到行业顶尖水平。\n- **数据资源零散**：寻找高质量的人脸属性、3D 形态及防攻击数据集如同大海捞针，导致模型训练数据多样性不足，泛化能力差。\n- **功能覆盖不全**：在需要人脸去模糊、超分辨率或活体检测等进阶功能时，找不到经过验证的开源库，被迫重复造轮子。\n\n### 使用 awesome-face 后\n- **快速锁定方案**：直接利用其整理的 CVPR 最新成果（如 VirFace、MagFace），迅速锁定了在遮挡和低画质下表现最强的检测与识别基线。\n- **对标行业顶尖**：通过集成的 PapersWithCode 榜单，实时对比模型在 WFLW 和 MegaFace 数据集上的表现，确保算法精度始终处于第一梯队。\n- **一站式数据获取**：按类别索引快速获取人脸 landmark、属性及对抗攻击等专用数据集，大幅缩短了数据清洗与预处理周期。\n- **全链路技术支撑**：从基础检测到高阶的换脸、老化模拟及防伪攻击，直接调用成熟的库与工具，将原本两个月的研发周期压缩至两周。\n\nawesome-face 通过聚合全球最前沿的人脸技术资源，帮助团队从繁琐的调研中解放出来，专注于核心业务逻辑的创新与落地。","https:\u002F\u002Foss.gittoolsai.com\u002Fimages\u002FbecauseofAI_awesome-face_36207a4c.png","becauseofAI",null,"https:\u002F\u002Foss.gittoolsai.com\u002Favatars\u002FbecauseofAI_64dc084a.jpg","helloai777@gmail.com","https:\u002F\u002Fgithub.com\u002FbecauseofAI",[81,85,89],{"name":82,"color":83,"percentage":84},"HTML","#e34c26",99.5,{"name":86,"color":87,"percentage":88},"CSS","#663399",0.3,{"name":90,"color":91,"percentage":92},"JavaScript","#f1e05a",0.2,1299,213,"2026-03-22T01:16:05","Apache-2.0",5,"","未说明",{"notes":101,"python":99,"dependencies":102},"该仓库（HelloFace\u002Fawesome-face）是一个人脸技术论文、数据集和工具的精选列表（Awesome List），而非一个可直接运行的单一软件工具。README 中列出了大量不同研究方向（如检测、识别、GAN、3D 重建等）的独立项目链接。因此，具体的运行环境需求（操作系统、GPU、内存、依赖库等）取决于用户选择克隆和运行的子项目。建议访问仓库中列出的具体项目页面以获取相应的安装指南。",[],[14,54,13],[67,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123],"face-paper","face-code","face-benchmark","face-recognition","face-detection","face-landmark","face-clustering","face-expression","face-action","face-3d","face-gan","face-deblurring","face-super-resolution","face-manipulation","face-anti-spoofing","machine-learning","deep-learning","artificial-intelligence","face","2026-03-27T02:49:30.150509","2026-04-06T05:17:31.297987",[],[]]