{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/mlfw-a-database-for-face-recognition-on","title":"MLFW: A Database for Face Recognition on Masked Faces","arxiv_id":"2109.05804","date":"2021-09-13","proceeding":null,"authors":["Chengrui Wang","Han Fang","Yaoyao Zhong","Weihong Deng"],"abstract":"As more and more people begin to wear masks due to current COVID-19 pandemic, existing face recognition systems may encounter severe performance degradation when recognizing masked faces. To figure out the impact of masks on face recognition model, we build a simple but effective tool to generate masked faces from unmasked faces automatically, and construct a new database called Masked LFW (MLFW) based on Cross-Age LFW (CALFW) database. The mask on the masked face generated by our method has good visual consistency with the original face. Moreover, we collect various mask templates, covering most of the common styles appeared in the daily life, to achieve diverse generation effects. Considering realistic scenarios, we design three kinds of combinations of face pairs. The recognition accuracy of SOTA models declines 5%-16% on MLFW database compared with the accuracy on the original images. MLFW database can be viewed and downloaded at \\url{http://whdeng.cn/mlfw}.","url_abs":"https://arxiv.org/abs/2109.05804v2","url_pdf":"https://arxiv.org/pdf/2109.05804v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"face-recognition","task_name":"Face Recognition"}],"methods":[],"datasets_introduced":[{"slug":"mlfw","name":"MLFW","full_name":"Masked LFW"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/face-recognition-on-mlfw","task":"Face Recognition","dataset":"MLFW","model":"MS1MV2, R100, SFace","rank_in_archive_order":1,"of":6,"metrics":{"Accuracy":"91.57"},"uses_additional_data":false},{"leaderboard":"/sota/face-recognition-on-mlfw","task":"Face Recognition","dataset":"MLFW","model":"MS1MV2, R100, Arcface","rank_in_archive_order":2,"of":6,"metrics":{"Accuracy":"90.57"},"uses_additional_data":false},{"leaderboard":"/sota/face-recognition-on-mlfw","task":"Face Recognition","dataset":"MLFW","model":"MS1MV2, R100, Curricularface","rank_in_archive_order":3,"of":6,"metrics":{"Accuracy":"90.43"},"uses_additional_data":false},{"leaderboard":"/sota/face-recognition-on-mlfw","task":"Face Recognition","dataset":"MLFW","model":"VGGFace2, R50, ArcFace","rank_in_archive_order":4,"of":6,"metrics":{"Accuracy":"85.95"},"uses_additional_data":false},{"leaderboard":"/sota/face-recognition-on-mlfw","task":"Face Recognition","dataset":"MLFW","model":"CASIA-WebFace, R50, CosFace","rank_in_archive_order":5,"of":6,"metrics":{"Accuracy":"82.52"},"uses_additional_data":false},{"leaderboard":"/sota/face-recognition-on-mlfw","task":"Face Recognition","dataset":"MLFW","model":"Private-Asia, R50, ArcFace","rank_in_archive_order":6,"of":6,"metrics":{"Accuracy":"77.20"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2109.05804","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}