{"id":17960223,"url":"https://github.com/cleardusk/mfr","last_synced_at":"2026-02-21T06:05:24.483Z","repository":{"id":89235275,"uuid":"247988009","full_name":"cleardusk/MFR","owner":"cleardusk","description":"Learning Meta Face Recognition in Unseen Domains, CVPR, Oral, 2020","archived":false,"fork":false,"pushed_at":"2021-12-29T09:19:06.000Z","size":970,"stargazers_count":149,"open_issues_count":5,"forks_count":16,"subscribers_count":18,"default_branch":"master","last_synced_at":"2025-10-19T10:46:18.408Z","etag":null,"topics":["cvpr-2020","domain-adaptation","domain-generalization","face-recognition","meta-learning"],"latest_commit_sha":null,"homepage":null,"language":null,"has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/cleardusk.png","metadata":{"files":{"readme":"readme.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null}},"created_at":"2020-03-17T14:16:22.000Z","updated_at":"2025-03-23T17:33:22.000Z","dependencies_parsed_at":"2023-06-14T13:15:36.231Z","dependency_job_id":null,"html_url":"https://github.com/cleardusk/MFR","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/cleardusk/MFR","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cleardusk%2FMFR","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cleardusk%2FMFR/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cleardusk%2FMFR/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cleardusk%2FMFR/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/cleardusk","download_url":"https://codeload.github.com/cleardusk/MFR/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/cleardusk%2FMFR/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":29674940,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-02-21T05:54:28.202Z","status":"ssl_error","status_checked_at":"2026-02-21T05:53:42.585Z","response_time":107,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["cvpr-2020","domain-adaptation","domain-generalization","face-recognition","meta-learning"],"created_at":"2024-10-29T11:05:43.580Z","updated_at":"2026-02-21T06:05:24.460Z","avatar_url":"https://github.com/cleardusk.png","language":null,"funding_links":[],"categories":[],"sub_categories":[],"readme":"# Learning Meta Face Recognition in Unseen Domains\n\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)\n\nBy [Jianzhu Guo](https://me.guojianzhu.com), [Xiangyu Zhu](http://www.cbsr.ia.ac.cn/users/xiangyuzhu/), [Chenxu Zhao](https://www.linkedin.com/in/chenxu-zhao-b66844107/), Dong Cao, [Zhen Lei](http://www.cbsr.ia.ac.cn/users/zlei/) and [Stan Z. Li](http://www.cbsr.ia.ac.cn/users/szli/).\n\n**\\[Todo\\]**\n\n- [ ] Release the images of the GFR-R and GFR-V benchmark.\n- [ ] Release the evaluation code.\n\n## Introduction\n\nThis repo will release the two proposed benchmarks GFR-R and GFR-V in our paper. I hope the proposed benchmarks will attract researchers on _Generalized Face Recognition_ problem. More details can be referred to our paper [Learning Meta Face Recognition in Unseen Domains](https://me.guojianzhu.com/assets/pdfs/05997.pdf), accepted to CVPR2020.\n\n\n### GFR Problem\n_What is GFR (Generalized Face Recognition) problem?_\n\nIn real-world applications of face recognition, the model trained on source domains D\u003csub\u003eS\u003c/sub\u003e is usually deployed in another domain D\u003csub\u003eT\u003c/sub\u003e with a different distribution. If the target domain D\u003csub\u003eT\u003c/sub\u003e is known and the data is accessible, it is categorized into domain adaptation for face recognition. If the target domain is unseen, it can be regarded as domain generalization for face recognition, and we call it Generalized Face Recognition, which is more common as the trained model is usually deployed in unknown scenarios and faced with unseen data. As shown below, the deployed model should be able to generalize to unseen domains without any updating or fine-tuning.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"imgs/GFR.jpg\" alt=\"bounding box\" width=\"640px\"\u003e\n\u003c/p\u003e\n\n### GFR-R and GFR-V Benchmark\nGFR-R is for crossing race evaluation and GFR-V is crossing facial variety, which means a large gap between source domains and target unseen domains. The involved datasets and protocols are shown in two tables below.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"imgs/GFR-datasets.png\" alt=\"bounding box\" width=\"800px\"\u003e\n\u003c/p\u003e\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"imgs/GFR-protocols.png\" alt=\"bounding box\" height=\"300px\"\u003e\n\u003c/p\u003e\n\n### Our Method\nOur proposed method are inspired by MAML, shown in the figure below.\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"imgs/method.jpg\" alt=\"bounding box\" width=\"960px\"\u003e\n\u003c/p\u003e\n\nThe meta-optimization procedure:\n\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"imgs/meta-opt.jpg\" alt=\"bounding box\" width=\"400\"\u003e\n\u003c/p\u003e\n\nPlease see [our paper](https://me.guojianzhu.com/assets/pdfs/05997.pdf) for more details.\n\n## Citation\n\n    @article{guo2020learning,\n        title   =   {Learning Meta Face Recognition in Unseen Domains},\n        author  =   {Guo, Jianzhu and Zhu, Xiangyu and Zhao, Chenxu and Cao, Dong and Lei, Zhen and Li, Stan Z},\n        journal =   {arXiv preprint arXiv:2003.07733},\n        year    =   {2020}\n    }\n\n\u003c!-- **This repo will keep updating.** --\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcleardusk%2Fmfr","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcleardusk%2Fmfr","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcleardusk%2Fmfr/lists"}