{"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/wider-face-a-face-detection-benchmark","title":"WIDER FACE: A Face Detection Benchmark","arxiv_id":"1511.06523","date":"2015-11-20","proceeding":"CVPR 2016 6","authors":["Shuo Yang","Ping Luo","Chen Change Loy","Xiaoou Tang"],"abstract":"Face detection is one of the most studied topics in the computer vision\ncommunity. Much of the progresses have been made by the availability of face\ndetection benchmark datasets. We show that there is a gap between current face\ndetection performance and the real world requirements. To facilitate future\nface detection research, we introduce the WIDER FACE dataset, which is 10 times\nlarger than existing datasets. The dataset contains rich annotations, including\nocclusions, poses, event categories, and face bounding boxes. Faces in the\nproposed dataset are extremely challenging due to large variations in scale,\npose and occlusion, as shown in Fig. 1. Furthermore, we show that WIDER FACE\ndataset is an effective training source for face detection. We benchmark\nseveral representative detection systems, providing an overview of\nstate-of-the-art performance and propose a solution to deal with large scale\nvariation. Finally, we discuss common failure cases that worth to be further\ninvestigated. Dataset can be downloaded at:\nmmlab.ie.cuhk.edu.hk/projects/WIDERFace","url_abs":"http://arxiv.org/abs/1511.06523v1","url_pdf":"http://arxiv.org/pdf/1511.06523v1.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":[{"paper_slug":"wider-face-a-face-detection-benchmark","repo_url":"https://github.com/kabrau/FaceDetection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"face-detection","task_name":"Face Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/face-detection-on-wider-face-hard","task":"Face Detection","dataset":"WIDER Face (Hard)","model":"Multiscale Cascade CNN","rank_in_archive_order":37,"of":40,"metrics":{"AP":"0.400"},"uses_additional_data":true},{"leaderboard":"/sota/face-detection-on-wider-face-hard","task":"Face Detection","dataset":"WIDER Face (Hard)","model":"Faceness-WIDER","rank_in_archive_order":38,"of":40,"metrics":{"AP":"0.315"},"uses_additional_data":false},{"leaderboard":"/sota/face-detection-on-wider-face-hard","task":"Face Detection","dataset":"WIDER Face (Hard)","model":"Two-stage CNN","rank_in_archive_order":39,"of":40,"metrics":{"AP":"0.304"},"uses_additional_data":false},{"leaderboard":"/sota/face-detection-on-wider-face-medium","task":"Face Detection","dataset":"WIDER Face (Medium)","model":"Multiscale Cascade CNN","rank_in_archive_order":34,"of":37,"metrics":{"AP":"0.636"},"uses_additional_data":false},{"leaderboard":"/sota/face-detection-on-wider-face-medium","task":"Face Detection","dataset":"WIDER Face (Medium)","model":"Faceness-WIDER","rank_in_archive_order":35,"of":37,"metrics":{"AP":"0.604"},"uses_additional_data":false},{"leaderboard":"/sota/face-detection-on-wider-face-medium","task":"Face Detection","dataset":"WIDER Face (Medium)","model":"Two-stage CNN","rank_in_archive_order":36,"of":37,"metrics":{"AP":"0.589"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1511.06523","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}