{"url":"/task/face-recognition","name":"Face Recognition","slug":"face-recognition","description_markdown":"**Facial Recognition** is the task of making a positive identification of a face in a photo or video image against a pre-existing database of faces. It begins with detection - distinguishing human faces from other objects in the image - and then works on identification of those detected faces.\r\n\r\nThe state of the art tables for this task are contained mainly in the consistent parts of the task : the face verification and face identification tasks.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Face Verification](https://shuftipro.com/face-verification) )</span>","categories":[{"name":"Computer Vision","url":"/area/computer-vision"},{"name":"Methodology","url":"/area/methodology"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":2329,"papers_with_code":639,"benchmarks":25,"benchmark_tables_in_archive":25,"benchmark_tables_shown":25,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":67,"subtasks":6,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/face-recognition-on-lfw","slug":"face-recognition-on-lfw","dataset":"LFW","dataset_url":"/dataset/lfw","rows_in_archive":16,"metrics":["Accuracy","FNMR [%] @ 10-3 FMR","F1-score","Precision","Recall"],"first_row_in_archive_order":{"model":"GhostFaceNetV2-1 (MS1MV3)","paper_title":"GhostFaceNets: Lightweight Face Recognition Model From Cheap Operations","paper_url":"/paper/ghostfacenets-lightweight-face-recognition","paper_date":"2023-04-10","arxiv_id":null,"code_links":[{"title":"serengil/deepface","url":"https://github.com/serengil/deepface"},{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"HamadYA/GhostFaceNets","url":"https://github.com/HamadYA/GhostFaceNets"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection","url":"https://github.com/Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection"},{"title":"Hazqeel09/ellzaf_ml","url":"https://github.com/Hazqeel09/ellzaf_ml"},{"title":"Armxyz1/Results-on-RFW","url":"https://github.com/Armxyz1/Results-on-RFW"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-cfp-fp","slug":"face-recognition-on-cfp-fp","dataset":"CFP-FP","dataset_url":null,"rows_in_archive":8,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"GhostFaceNetV2-1","paper_title":"GhostFaceNets: Lightweight Face Recognition Model From Cheap Operations","paper_url":"/paper/ghostfacenets-lightweight-face-recognition","paper_date":"2023-04-10","arxiv_id":null,"code_links":[{"title":"serengil/deepface","url":"https://github.com/serengil/deepface"},{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"HamadYA/GhostFaceNets","url":"https://github.com/HamadYA/GhostFaceNets"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection","url":"https://github.com/Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection"},{"title":"Hazqeel09/ellzaf_ml","url":"https://github.com/Hazqeel09/ellzaf_ml"},{"title":"Armxyz1/Results-on-RFW","url":"https://github.com/Armxyz1/Results-on-RFW"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-casia-webface-masks","slug":"face-recognition-on-casia-webface-masks","dataset":"CASIA-WebFace+masks","dataset_url":"/dataset/casia-webface-masks","rows_in_archive":6,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"Fine-tuned ArcFace","paper_title":"A realistic approach to generate masked faces applied on two novel masked face recognition data sets","paper_url":"/paper/a-realistic-approach-to-generate-masked-faces","paper_date":"2021-09-03","arxiv_id":"2109.01745","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"securifai/masked_faces","url":"https://github.com/securifai/masked_faces"}],"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":1}}},{"leaderboard":"/sota/face-recognition-on-celeba-masks","slug":"face-recognition-on-celeba-masks","dataset":"CelebA+masks","dataset_url":"/dataset/celeba-masks","rows_in_archive":6,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"Fine-tuned ArcFace","paper_title":"A realistic approach to generate masked faces applied on two novel masked face recognition data sets","paper_url":"/paper/a-realistic-approach-to-generate-masked-faces","paper_date":"2021-09-03","arxiv_id":"2109.01745","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"securifai/masked_faces","url":"https://github.com/securifai/masked_faces"}],"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":1}}},{"leaderboard":"/sota/face-recognition-on-mlfw","slug":"face-recognition-on-mlfw","dataset":"MLFW","dataset_url":"/dataset/mlfw","rows_in_archive":6,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"MS1MV2, R100, SFace","paper_title":"MLFW: A Database for Face Recognition on Masked Faces","paper_url":"/paper/mlfw-a-database-for-face-recognition-on","paper_date":"2021-09-13","arxiv_id":"2109.05804","code_links":[],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-agedb-30","slug":"face-recognition-on-agedb-30","dataset":"AgeDB-30","dataset_url":null,"rows_in_archive":4,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"Prodpoly","paper_title":"Deep Polynomial Neural Networks","paper_url":"/paper/deep-polynomial-neural-networks","paper_date":"2020-06-20","arxiv_id":"2006.13026","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"grigorisg9gr/polynomial_nets","url":"https://github.com/grigorisg9gr/polynomial_nets"},{"title":"Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection","url":"https://github.com/Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection"},{"title":"jesperhauch/polynomial_deep_learning","url":"https://github.com/jesperhauch/polynomial_deep_learning"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-color-feret","slug":"face-recognition-on-color-feret","dataset":"Color FERET","dataset_url":"/dataset/color-feret","rows_in_archive":4,"metrics":["FNMR [%] @ 10-3 FMR","5-class test accuracy"],"first_row_in_archive_order":{"model":"PIC - QMagFace","paper_title":"PIC-Score: Probabilistic Interpretable Comparison Score for Optimal Matching Confidence in Single- and Multi-Biometric (Face) Recognition","paper_url":"/paper/pic-score-probabilistic-interpretable","paper_date":"2022-11-22","arxiv_id":"2211.12483","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"pterhoer/optimalmatchingconfidence","url":"https://github.com/pterhoer/optimalmatchingconfidence"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-ijb-b","slug":"face-recognition-on-ijb-b","dataset":"IJB-B","dataset_url":"/dataset/ijb-b","rows_in_archive":4,"metrics":["Rank-1","Rank-5","TAR @ FAR=0.0001","TAR @ FAR=1e-3","TAR @ FAR=1e-4","TAR @ FAR=1e-5"],"first_row_in_archive_order":{"model":"ArcFace+CSFM","paper_title":"Controllable and Guided Face Synthesis for Unconstrained Face Recognition","paper_url":"/paper/controllable-and-guided-face-synthesis-for","paper_date":"2022-07-20","arxiv_id":"2207.10180","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"liuf1990/CFSM","url":"https://github.com/liuf1990/CFSM"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-adience","slug":"face-recognition-on-adience","dataset":"Adience","dataset_url":"/dataset/adience","rows_in_archive":3,"metrics":["FNMR [%] @ 10-3 FMR"],"first_row_in_archive_order":{"model":"PIC - MagFace","paper_title":"PIC-Score: Probabilistic Interpretable Comparison Score for Optimal Matching Confidence in Single- and Multi-Biometric (Face) Recognition","paper_url":"/paper/pic-score-probabilistic-interpretable","paper_date":"2022-11-22","arxiv_id":"2211.12483","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"pterhoer/optimalmatchingconfidence","url":"https://github.com/pterhoer/optimalmatchingconfidence"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-calfw","slug":"face-recognition-on-calfw","dataset":"CALFW","dataset_url":"/dataset/calfw","rows_in_archive":3,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"Prodpoly","paper_title":"Deep Polynomial Neural Networks","paper_url":"/paper/deep-polynomial-neural-networks","paper_date":"2020-06-20","arxiv_id":"2006.13026","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"grigorisg9gr/polynomial_nets","url":"https://github.com/grigorisg9gr/polynomial_nets"},{"title":"Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection","url":"https://github.com/Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection"},{"title":"jesperhauch/polynomial_deep_learning","url":"https://github.com/jesperhauch/polynomial_deep_learning"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-mfw","slug":"face-recognition-on-mfw","dataset":"MFW+ (U-M)","dataset_url":"/dataset/mfw","rows_in_archive":3,"metrics":["TAR@FAR=0.0001"],"first_row_in_archive_order":{"model":"AdaFace+PPL","paper_title":"PPL: Pairwise Prototype Learning for Masked Face Recognition","paper_url":"/paper/ppl-pairwise-prototype-learning-for-masked","paper_date":"2022-11-21","arxiv_id":null,"code_links":[{"title":"kim1102/PPL-MFR","url":"https://github.com/kim1102/PPL-MFR"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-mfw-m-m","slug":"face-recognition-on-mfw-m-m","dataset":"MFW+ (M-M)","dataset_url":"/dataset/mfw-m-m","rows_in_archive":3,"metrics":["TAR@FAR=0.0001"],"first_row_in_archive_order":{"model":"ArcFace + PPL","paper_title":"PPL: Pairwise Prototype Learning for Masked Face Recognition","paper_url":"/paper/ppl-pairwise-prototype-learning-for-masked","paper_date":"2022-11-21","arxiv_id":null,"code_links":[{"title":"kim1102/PPL-MFR","url":"https://github.com/kim1102/PPL-MFR"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-morph","slug":"face-recognition-on-morph","dataset":"MORPH","dataset_url":"/dataset/morph","rows_in_archive":3,"metrics":["FNMR [%] @ 10-3 FMR"],"first_row_in_archive_order":{"model":"PIC - ArcFace","paper_title":"PIC-Score: Probabilistic Interpretable Comparison Score for Optimal Matching Confidence in Single- and Multi-Biometric (Face) Recognition","paper_url":"/paper/pic-score-probabilistic-interpretable","paper_date":"2022-11-22","arxiv_id":"2211.12483","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"pterhoer/optimalmatchingconfidence","url":"https://github.com/pterhoer/optimalmatchingconfidence"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-adience-online-open-set","slug":"face-recognition-on-adience-online-open-set","dataset":"Adience (Online Open Set)","dataset_url":"/dataset/adience","rows_in_archive":2,"metrics":["Average Accuracy (10 times)"],"first_row_in_archive_order":{"model":"FaceNet+Adaptive Threshold","paper_title":"Data-specific Adaptive Threshold for Face Recognition and Authentication","paper_url":"/paper/data-specific-adaptive-threshold-for-face","paper_date":"2018-10-26","arxiv_id":"1810.11160","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"ivclab/Online-Face-Recognition-and-Authentication","url":"https://github.com/ivclab/Online-Face-Recognition-and-Authentication"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-carl","slug":"face-recognition-on-carl","dataset":"Carl","dataset_url":"/dataset/carl-1","rows_in_archive":2,"metrics":["Rank-1"],"first_row_in_archive_order":{"model":"Model with Up Convolution + DoG Filter (Aligned)","paper_title":"Thermal to Visible Face Recognition Using Deep Autoencoders","paper_url":"/paper/thermal-to-visible-face-recognition-using","paper_date":"2020-02-10","arxiv_id":"2002.04219","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"Alpkant/Thermal-to-Visible-Face-Recognition-Using-Deep-Autoencoders","url":"https://github.com/Alpkant/Thermal-to-Visible-Face-Recognition-Using-Deep-Autoencoders"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-color-feret-online-open","slug":"face-recognition-on-color-feret-online-open","dataset":"Color FERET (Online Open Set)","dataset_url":"/dataset/color-feret","rows_in_archive":2,"metrics":["Average Accuracy (10 times)"],"first_row_in_archive_order":{"model":"FaceNet+Adaptive Threshold","paper_title":"Data-specific Adaptive Threshold for Face Recognition and Authentication","paper_url":"/paper/data-specific-adaptive-threshold-for-face","paper_date":"2018-10-26","arxiv_id":"1810.11160","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"ivclab/Online-Face-Recognition-and-Authentication","url":"https://github.com/ivclab/Online-Face-Recognition-and-Authentication"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-cplfw","slug":"face-recognition-on-cplfw","dataset":"CPLFW","dataset_url":"/dataset/cplfw","rows_in_archive":2,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"GhostFaceNetV2-1","paper_title":"GhostFaceNets: Lightweight Face Recognition Model From Cheap Operations","paper_url":"/paper/ghostfacenets-lightweight-face-recognition","paper_date":"2023-04-10","arxiv_id":null,"code_links":[{"title":"serengil/deepface","url":"https://github.com/serengil/deepface"},{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"HamadYA/GhostFaceNets","url":"https://github.com/HamadYA/GhostFaceNets"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection","url":"https://github.com/Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection"},{"title":"Hazqeel09/ellzaf_ml","url":"https://github.com/Hazqeel09/ellzaf_ml"},{"title":"Armxyz1/Results-on-RFW","url":"https://github.com/Armxyz1/Results-on-RFW"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-lfw-online-open-set","slug":"face-recognition-on-lfw-online-open-set","dataset":"LFW (Online Open Set)","dataset_url":"/dataset/lfw","rows_in_archive":2,"metrics":["Average Accuracy (10 times)"],"first_row_in_archive_order":{"model":"FaceNet+Adaptive Threshold","paper_title":"Data-specific Adaptive Threshold for Face Recognition and Authentication","paper_url":"/paper/data-specific-adaptive-threshold-for-face","paper_date":"2018-10-26","arxiv_id":"1810.11160","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"ivclab/Online-Face-Recognition-and-Authentication","url":"https://github.com/ivclab/Online-Face-Recognition-and-Authentication"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-uhdb31","slug":"face-recognition-on-uhdb31","dataset":"UHDB31","dataset_url":null,"rows_in_archive":2,"metrics":["Rank-1"],"first_row_in_archive_order":{"model":"Wang et al. [5]","paper_title":"Finding Missing Children: Aging Deep Face Features","paper_url":"/paper/aging-deep-face-features-finding-missing","paper_date":"2019-11-18","arxiv_id":"1911.07538","code_links":[],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-und-x1","slug":"face-recognition-on-und-x1","dataset":"UND-X1","dataset_url":null,"rows_in_archive":2,"metrics":["Rank-1"],"first_row_in_archive_order":{"model":"Model with Up Convolution + DoG Filter","paper_title":"Thermal to Visible Face Recognition Using Deep Autoencoders","paper_url":"/paper/thermal-to-visible-face-recognition-using","paper_date":"2020-02-10","arxiv_id":"2002.04219","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"Alpkant/Thermal-to-Visible-Face-Recognition-Using-Deep-Autoencoders","url":"https://github.com/Alpkant/Thermal-to-Visible-Face-Recognition-Using-Deep-Autoencoders"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-bts3-1","slug":"face-recognition-on-bts3-1","dataset":"BTS3.1","dataset_url":"/dataset/bts3-1","rows_in_archive":1,"metrics":["TAR @ FAR=0.01"],"first_row_in_archive_order":{"model":"MCN","paper_title":"Multicolumn Networks for Face Recognition","paper_url":"/paper/multicolumn-networks-for-face-recognition","paper_date":"2018-07-24","arxiv_id":"1807.09192","code_links":[{"title":"ibendrup/MulticolumnNetwork","url":"https://github.com/ibendrup/MulticolumnNetwork"}],"syntology":{"n":4,"n_ran":0,"n_unverified":4,"n_pointer_only":0}}},{"leaderboard":"/sota/face-recognition-on-cfp-ff","slug":"face-recognition-on-cfp-ff","dataset":"CFP-FF","dataset_url":null,"rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"GhostFaceNetV2-1","paper_title":"GhostFaceNets: Lightweight Face Recognition Model From Cheap Operations","paper_url":"/paper/ghostfacenets-lightweight-face-recognition","paper_date":"2023-04-10","arxiv_id":null,"code_links":[{"title":"serengil/deepface","url":"https://github.com/serengil/deepface"},{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"HamadYA/GhostFaceNets","url":"https://github.com/HamadYA/GhostFaceNets"},{"title":"FaceOnLive/Face-Recognition-SDK-Android","url":"https://github.com/FaceOnLive/Face-Recognition-SDK-Android"},{"title":"Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection","url":"https://github.com/Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection"},{"title":"Hazqeel09/ellzaf_ml","url":"https://github.com/Hazqeel09/ellzaf_ml"},{"title":"Armxyz1/Results-on-RFW","url":"https://github.com/Armxyz1/Results-on-RFW"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-eurecom","slug":"face-recognition-on-eurecom","dataset":"EURECOM","dataset_url":null,"rows_in_archive":1,"metrics":["Rank-1"],"first_row_in_archive_order":{"model":"Model with Up Convolution + DoG Filter","paper_title":"Thermal to Visible Face Recognition Using Deep Autoencoders","paper_url":"/paper/thermal-to-visible-face-recognition-using","paper_date":"2020-02-10","arxiv_id":"2002.04219","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"Alpkant/Thermal-to-Visible-Face-Recognition-Using-Deep-Autoencoders","url":"https://github.com/Alpkant/Thermal-to-Visible-Face-Recognition-Using-Deep-Autoencoders"}],"syntology":null}},{"leaderboard":"/sota/face-recognition-on-mfr","slug":"face-recognition-on-mfr","dataset":"MFR","dataset_url":"/dataset/mfr","rows_in_archive":1,"metrics":["MFR-ALL","MFR-MASK","African","Caucasian","South Asian","East Asian"],"first_row_in_archive_order":{"model":"Partial FC","paper_title":"Killing Two Birds with One Stone:Efficient and Robust Training of Face Recognition CNNs by Partial FC","paper_url":"/paper/killing-two-birds-with-one-stone-efficient","paper_date":"2022-03-28","arxiv_id":"2203.15565","code_links":[{"title":"deepinsight/insightface","url":"https://github.com/deepinsight/insightface"},{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection","url":"https://github.com/Recognito-Vision/Android-FaceRecognition-FaceLivenessDetection"},{"title":"pedrobvidal/insightface","url":"https://github.com/pedrobvidal/insightface"},{"title":"bovifocr/insightface","url":"https://github.com/bovifocr/insightface"},{"title":"yangyucheng000/-insightface","url":"https://github.com/yangyucheng000/-insightface"}],"syntology":{"n":12,"n_ran":4,"n_unverified":8,"n_pointer_only":4}}},{"leaderboard":"/sota/face-recognition-on-xqlfw","slug":"face-recognition-on-xqlfw","dataset":"XQLFW","dataset_url":"/dataset/xqlfw","rows_in_archive":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"FaceTransformer+OctupletLoss","paper_title":"Octuplet Loss: Make Face Recognition Robust to Image Resolution","paper_url":"/paper/octuplet-loss-make-face-recognition-robust-to","paper_date":"2022-07-14","arxiv_id":"2207.06726","code_links":[{"title":"Faceplugin-ltd/FaceRecognition-Android","url":"https://github.com/Faceplugin-ltd/FaceRecognition-Android"},{"title":"martlgap/octuplet-loss","url":"https://github.com/martlgap/octuplet-loss"}],"syntology":null}}],"datasets":[{"url":"/dataset/lfw","name":"LFW","full_name":"Labeled Faces in the Wild","num_papers_in_archive":820},{"url":"/dataset/casia-webface","name":"CASIA-WebFace","full_name":"","num_papers_in_archive":415},{"url":"/dataset/ms-celeb-1m","name":"MS-Celeb-1M","full_name":"","num_papers_in_archive":257},{"url":"/dataset/extended-yale-b-1","name":"Extended Yale B","full_name":"","num_papers_in_archive":185},{"url":"/dataset/morph","name":"MORPH","full_name":"","num_papers_in_archive":180},{"url":"/dataset/ijb-b","name":"IJB-B","full_name":"IARPA Janus Benchmark-B","num_papers_in_archive":163},{"url":"/dataset/vgg-face-1","name":"VGG Face","full_name":"","num_papers_in_archive":94},{"url":"/dataset/rfw","name":"RFW","full_name":"Racial Faces in-the-Wild","num_papers_in_archive":73},{"url":"/dataset/casia-fasd","name":"CASIA-FASD","full_name":"CASIA-FASD","num_papers_in_archive":42},{"url":"/dataset/partial-reid","name":"Partial-REID","full_name":"Partial-REID","num_papers_in_archive":41},{"url":"/dataset/color-feret","name":"Color FERET","full_name":"Color FERET","num_papers_in_archive":36},{"url":"/dataset/adience","name":"Adience","full_name":null,"num_papers_in_archive":35},{"url":"/dataset/umdfaces","name":"UMDFaces","full_name":"","num_papers_in_archive":34},{"url":"/dataset/tinyface","name":"TinyFace","full_name":"","num_papers_in_archive":32},{"url":"/dataset/celeba-spoof","name":"CelebA-Spoof","full_name":"","num_papers_in_archive":28},{"url":"/dataset/casia-surf","name":"CASIA-SURF","full_name":"","num_papers_in_archive":24},{"url":"/dataset/imdb-face","name":"IMDb-Face","full_name":"","num_papers_in_archive":22},{"url":"/dataset/digiface-1m","name":"DigiFace-1M","full_name":"","num_papers_in_archive":21},{"url":"/dataset/webface260m","name":"WebFace260M","full_name":"","num_papers_in_archive":20},{"url":"/dataset/replay-mobile-1","name":"Replay-Mobile","full_name":"Replay-Mobile","num_papers_in_archive":18},{"url":"/dataset/calfw","name":"CALFW","full_name":"Cross-Age LFW","num_papers_in_archive":17},{"url":"/dataset/cplfw","name":"CPLFW","full_name":"Cross-Pose LFW","num_papers_in_archive":17},{"url":"/dataset/mfr","name":"MFR","full_name":"Ongoing version of ICCV-2021 Masked Face Recognition Challenge & Workshop(MFR)","num_papers_in_archive":16},{"url":"/dataset/xqlfw","name":"XQLFW","full_name":"Cross-Quality Labeled Faces in the Wild","num_papers_in_archive":16},{"url":"/dataset/idifface","name":"IDiff-Face","full_name":"","num_papers_in_archive":13},{"url":"/dataset/diveface","name":"DiveFace","full_name":"","num_papers_in_archive":11},{"url":"/dataset/rmfd","name":"RMFD","full_name":"Real-World Masked Face Dataset","num_papers_in_archive":11},{"url":"/dataset/demogpairs","name":"DemogPairs","full_name":"","num_papers_in_archive":10},{"url":"/dataset/qmul-survface","name":"QMUL-SurvFace","full_name":"","num_papers_in_archive":10},{"url":"/dataset/icartoonface","name":"iCartoonFace","full_name":null,"num_papers_in_archive":9},{"url":"/dataset/meglass","name":"MeGlass","full_name":null,"num_papers_in_archive":8},{"url":"/dataset/dfw","name":"DFW","full_name":"Disguised Faces in the Wild","num_papers_in_archive":7},{"url":"/dataset/dronesurf","name":"DroneSURF","full_name":"DroneSURF: Benchmark Dataset for Drone-based Face Recognition","num_papers_in_archive":6},{"url":"/dataset/iqiyi-vid","name":"iQIYI-VID","full_name":"iQIYI-VID","num_papers_in_archive":6},{"url":"/dataset/mcxface","name":"MCXFACE","full_name":"Multi-Channel Heterogeneous Face Recognition dataset","num_papers_in_archive":6},{"url":"/dataset/mebal","name":"mEBAL","full_name":"","num_papers_in_archive":6},{"url":"/dataset/mlfw","name":"MLFW","full_name":"Masked LFW","num_papers_in_archive":6},{"url":"/dataset/webcaricature-dataset","name":"WebCaricature Dataset","full_name":"","num_papers_in_archive":6},{"url":"/dataset/bts3-1","name":"BTS3.1","full_name":"Expanding Accurate Person Recognition to New Altitudes and Ranges: The BRIAR Dataset","num_papers_in_archive":5},{"url":"/dataset/rof","name":"ROF","full_name":"Real World Occluded Faces","num_papers_in_archive":5},{"url":"/dataset/1995-people-face-images-data-asian-race","name":"1,995 People Face Images Data (Asian race)","full_name":"1,995 People Face Images Data (Asian race)","num_papers_in_archive":4},{"url":"/dataset/casia-webface-masks","name":"CASIA-WebFace+masks","full_name":"","num_papers_in_archive":4},{"url":"/dataset/celeba-masks","name":"CelebA+masks","full_name":"","num_papers_in_archive":4},{"url":"/dataset/casia-face-africa","name":"CASIA-Face-Africa","full_name":"","num_papers_in_archive":3},{"url":"/dataset/ijb-s","name":"IJB-S","full_name":"IARPA Janus Benchmark-S","num_papers_in_archive":3},{"url":"/dataset/imdb-clean","name":"IMDB-Clean","full_name":"","num_papers_in_archive":3},{"url":"/dataset/5011-images-human-frontal-face-data-male","name":"5,011 Images – Human Frontal face Data (Male)","full_name":"5,011 Images – Human Frontal face Data (Male)","num_papers_in_archive":2},{"url":"/dataset/celeb-hq-face-gender-recognition-dataset","name":"Celeb-HQ Face Gender Recognition Dataset","full_name":"Celeb-HQ Face Gender Recognition Dataset","num_papers_in_archive":2},{"url":"/dataset/celeb-hq-facial-identity-recognition-dataset","name":"Celeb-HQ Facial Identity Recognition Dataset","full_name":"Celeb-HQ Facial Identity Recognition Dataset","num_papers_in_archive":2},{"url":"/dataset/kanface","name":"KANFace","full_name":"KANFace Dataset","num_papers_in_archive":2},{"url":"/dataset/mfw","name":"MFW+ (U-M)","full_name":"","num_papers_in_archive":2},{"url":"/dataset/swax","name":"SWAX","full_name":"Sense Wax Attack dataset","num_papers_in_archive":2},{"url":"/dataset/23-pairs-of-identical-twins-face-image-data","name":"23 Pairs of Identical Twins Face Image Data","full_name":"23 Pairs of Identical Twins Face Image Data","num_papers_in_archive":1},{"url":"/dataset/curated-afd","name":"Curated AFD","full_name":null,"num_papers_in_archive":1},{"url":"/dataset/extended-youtube-faces-e-ytf","name":"Extended YouTube Faces (E-YTF)","full_name":"","num_papers_in_archive":1},{"url":"/dataset/fad","name":"FAD","full_name":"Face Attributes Dataset","num_papers_in_archive":1},{"url":"/dataset/imfw","name":"IMFW","full_name":"Indian Masked Faces In The Wild","num_papers_in_archive":1},{"url":"/dataset/chicago-face-database-cfd","name":"international faces","full_name":"","num_papers_in_archive":1},{"url":"/dataset/lslf","name":"LSLF","full_name":"Large-scale Labeled Face","num_papers_in_archive":1},{"url":"/dataset/ltft","name":"LTFT","full_name":"Long-Term Face Tracking","num_papers_in_archive":1},{"url":"/dataset/mfsd","name":"MFSD","full_name":"Masked Face Segmentation Dataset","num_papers_in_archive":1},{"url":"/dataset/mfw-m-m","name":"MFW+ (M-M)","full_name":"","num_papers_in_archive":1},{"url":"/dataset/ocfr-lfw","name":"OCFR-LFW","full_name":"Occluded Face Recognition LFW","num_papers_in_archive":1},{"url":"/dataset/uccs","name":"UCCS","full_name":"UnConstrained College Students Dataset","num_papers_in_archive":1},{"url":"/dataset/wildestfaces","name":"WildestFaces","full_name":"","num_papers_in_archive":1},{"url":"/dataset/yfcc-celeba","name":"YFCC-CelebA","full_name":"","num_papers_in_archive":1},{"url":"/dataset/facesinthings","name":"FacesInThings","full_name":"","num_papers_in_archive":0}],"subtasks":[{"url":"/task/age-invariant-face-recognition","name":"Age-Invariant Face Recognition"},{"url":"/task/face-image-quality-assessment","name":"Face Image Quality Assessment"},{"url":"/task/face-quality-assessement","name":"Face Quality Assessement"},{"url":"/task/lightweight-face-recognition","name":"Lightweight Face Recognition"},{"url":"/task/synthetic-face-recognition","name":"Synthetic Face Recognition"},{"url":"/task/unsupervised-face-recognition","name":"Unsupervised face recognition"}],"parent_tasks":[{"url":"/task/facial-recognition-and-modelling","name":"Facial Recognition and Modelling"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":639,"tagged_in_all":2329,"items":[{"url":"/paper/facenet-a-unified-embedding-for-face","title":"FaceNet: A Unified Embedding for Face Recognition and Clustering","date":"2015-03-12","arxiv_id":"1503.03832","repositories_listed":183,"syntology":{"n":154,"n_ran":65,"n_unverified":89,"n_pointer_only":45}},{"url":"/paper/arcface-additive-angular-margin-loss-for-deep","title":"ArcFace: Additive Angular Margin Loss for Deep Face Recognition","date":"2018-01-23","arxiv_id":"1801.07698","repositories_listed":100,"syntology":{"n":21,"n_ran":16,"n_unverified":5,"n_pointer_only":14}},{"url":"/paper/vggface2-a-dataset-for-recognising-faces","title":"VGGFace2: A dataset for recognising faces across pose and age","date":"2017-10-23","arxiv_id":"1710.08092","repositories_listed":24,"syntology":{"n":8,"n_ran":5,"n_unverified":3,"n_pointer_only":5}},{"url":"/paper/sphereface-deep-hypersphere-embedding-for","title":"SphereFace: Deep Hypersphere Embedding for Face Recognition","date":"2017-04-26","arxiv_id":"1704.08063","repositories_listed":22,"syntology":{"n":2,"n_ran":1,"n_unverified":1,"n_pointer_only":1}},{"url":"/paper/a-light-cnn-for-deep-face-representation-with","title":"A Light CNN for Deep Face Representation with Noisy Labels","date":"2015-11-09","arxiv_id":"1511.02683","repositories_listed":19,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":1}},{"url":"/paper/circle-loss-a-unified-perspective-of-pair","title":"Circle Loss: A Unified Perspective of Pair Similarity Optimization","date":"2020-02-25","arxiv_id":"2002.10857","repositories_listed":16,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":1}},{"url":"/paper/learning-face-representation-from-scratch","title":"Learning Face Representation from Scratch","date":"2014-11-28","arxiv_id":"1411.7923","repositories_listed":15,"syntology":{"n":11,"n_ran":1,"n_unverified":10,"n_pointer_only":0}},{"url":"/paper/ms-celeb-1m-a-dataset-and-benchmark-for-large","title":"MS-Celeb-1M: A Dataset and Benchmark for Large-Scale Face Recognition","date":"2016-07-27","arxiv_id":"1607.08221","repositories_listed":12,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":3}},{"url":"/paper/cosface-large-margin-cosine-loss-for-deep","title":"CosFace: Large Margin Cosine Loss for Deep Face Recognition","date":"2018-01-29","arxiv_id":"1801.09414","repositories_listed":11,"syntology":{"n":2,"n_ran":1,"n_unverified":1,"n_pointer_only":2}},{"url":"/paper/repmlp-re-parameterizing-convolutions-into","title":"RepMLP: Re-parameterizing Convolutions into Fully-connected Layers for Image Recognition","date":"2021-05-05","arxiv_id":"2105.01883","repositories_listed":10,"syntology":{"n":14,"n_ran":2,"n_unverified":12,"n_pointer_only":0}},{"url":"/paper/adaface-quality-adaptive-margin-for-face","title":"AdaFace: Quality Adaptive Margin for Face Recognition","date":"2022-04-03","arxiv_id":"2204.00964","repositories_listed":9,"syntology":{"n":22,"n_ran":17,"n_unverified":5,"n_pointer_only":11}},{"url":"/paper/can-we-still-avoid-automatic-face-detection","title":"Can we still avoid automatic face detection?","date":"2016-02-14","arxiv_id":"1602.04504","repositories_listed":9,"syntology":null},{"url":"/paper/deepid3-face-recognition-with-very-deep","title":"DeepID3: Face Recognition with Very Deep Neural Networks","date":"2015-02-03","arxiv_id":"1502.00873","repositories_listed":9,"syntology":null},{"url":"/paper/learning-meta-face-recognition-in-unseen","title":"Learning Meta Face Recognition in Unseen Domains","date":"2020-03-17","arxiv_id":"2003.07733","repositories_listed":8,"syntology":null},{"url":"/paper/deep-face-recognition-a-survey","title":"Deep Face Recognition: A Survey","date":"2018-04-18","arxiv_id":"1804.06655","repositories_listed":8,"syntology":null},{"url":"/paper/ghostfacenets-lightweight-face-recognition","title":"GhostFaceNets: Lightweight Face Recognition Model From Cheap Operations","date":"2023-04-10","arxiv_id":null,"repositories_listed":7,"syntology":null},{"url":"/paper/partial-fc-training-10-million-identities-on","title":"Partial FC: Training 10 Million Identities on a Single Machine","date":"2020-10-11","arxiv_id":"2010.05222","repositories_listed":7,"syntology":null},{"url":"/paper/deep-pixel-wise-binary-supervision-for-face","title":"Deep Pixel-wise Binary Supervision for Face Presentation Attack Detection","date":"2019-07-09","arxiv_id":"1907.04047","repositories_listed":7,"syntology":null},{"url":"/paper/faceqnet-quality-assessment-for-face","title":"FaceQnet: Quality Assessment for Face Recognition based on Deep Learning","date":"2019-04-03","arxiv_id":"1904.01740","repositories_listed":7,"syntology":null},{"url":"/paper/compact-bilinear-pooling","title":"Compact Bilinear Pooling","date":"2015-11-19","arxiv_id":"1511.06062","repositories_listed":7,"syntology":null},{"url":"/paper/sface-sigmoid-constrained-hypersphere-loss-1","title":"SFace: Sigmoid-Constrained Hypersphere Loss for Robust Face Recognition","date":"2022-05-24","arxiv_id":"2205.12010","repositories_listed":6,"syntology":null},{"url":"/paper/killing-two-birds-with-one-stone-efficient","title":"Killing Two Birds with One Stone:Efficient and Robust Training of Face Recognition CNNs by Partial FC","date":"2022-03-28","arxiv_id":"2203.15565","repositories_listed":6,"syntology":{"n":12,"n_ran":4,"n_unverified":8,"n_pointer_only":4}},{"url":"/paper/masked-face-recognition-for-secure","title":"Masked Face Recognition for Secure Authentication","date":"2020-08-25","arxiv_id":"2008.11104","repositories_listed":6,"syntology":{"n":12,"n_ran":1,"n_unverified":11,"n_pointer_only":0}},{"url":"/paper/seesawfacenets-sparse-and-robust-face","title":"SeesawFaceNets: sparse and robust face verification model for mobile platform","date":"2019-08-24","arxiv_id":"1908.09124","repositories_listed":6,"syntology":{"n":17,"n_ran":7,"n_unverified":10,"n_pointer_only":5}},{"url":"/paper/adacos-adaptively-scaling-cosine-logits-for","title":"AdaCos: Adaptively Scaling Cosine Logits for Effectively Learning Deep Face Representations","date":"2019-05-01","arxiv_id":"1905.00292","repositories_listed":6,"syntology":{"n":5,"n_ran":0,"n_unverified":5,"n_pointer_only":0}},{"url":"/paper/deep-polynomial-neural-networks","title":"Deep Polynomial Neural Networks","date":"2020-06-20","arxiv_id":"2006.13026","repositories_listed":5,"syntology":null},{"url":"/paper/ser-fiq-unsupervised-estimation-of-face-image","title":"SER-FIQ: Unsupervised Estimation of Face Image Quality Based on Stochastic Embedding Robustness","date":"2020-03-20","arxiv_id":"2003.09373","repositories_listed":5,"syntology":null},{"url":"/paper/deep-spatial-gradient-and-temporal-depth","title":"Deep Spatial Gradient and Temporal Depth Learning for Face Anti-spoofing","date":"2020-03-18","arxiv_id":"2003.08061","repositories_listed":5,"syntology":null},{"url":"/paper/searching-central-difference-convolutional","title":"Searching Central Difference Convolutional Networks for Face Anti-Spoofing","date":"2020-03-09","arxiv_id":"2003.04092","repositories_listed":5,"syntology":null},{"url":"/paper/support-vector-guided-softmax-loss-for-face","title":"Support Vector Guided Softmax Loss for Face Recognition","date":"2018-12-29","arxiv_id":"1812.11317","repositories_listed":5,"syntology":null}],"syntology_records":15,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}