{"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/adversarial-occlusion-aware-face-detection","title":"Adversarial Occlusion-aware Face Detection","arxiv_id":"1709.05188","date":"2017-09-15","proceeding":null,"authors":["Yujia Chen","Lingxiao Song","Ran He"],"abstract":"Occluded face detection is a challenging detection task due to the large\nappearance variations incurred by various real-world occlusions. This paper\nintroduces an Adversarial Occlusion-aware Face Detector (AOFD) by\nsimultaneously detecting occluded faces and segmenting occluded areas.\nSpecifically, we employ an adversarial training strategy to generate\nocclusion-like face features that are difficult for a face detector to\nrecognize. Occlusion mask is predicted simultaneously while detecting occluded\nfaces and the occluded area is utilized as an auxiliary instead of being\nregarded as a hindrance. Moreover, the supervisory signals from the\nsegmentation branch will reversely affect the features, aiding in detecting\nheavily-occluded faces accordingly. Consequently, AOFD is able to find the\nfaces with few exposed facial landmarks with very high confidences and keeps\nhigh detection accuracy even for masked faces. Extensive experiments\ndemonstrate that AOFD not only significantly outperforms state-of-the-art\nmethods on the MAFA occluded face detection dataset, but also achieves\ncompetitive detection accuracy on benchmark dataset for general face detection\nsuch as FDDB.","url_abs":"http://arxiv.org/abs/1709.05188v6","url_pdf":"http://arxiv.org/pdf/1709.05188v6.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":"adversarial-occlusion-aware-face-detection","repo_url":"https://github.com/IssacCyj/Adversarial-Occlussion-aware-Face-Detection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"caffe2","reach":null}],"tasks":[{"task_slug":"face-detection","task_name":"Face Detection"},{"task_slug":"occluded-face-detection","task_name":"Occluded Face Detection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/occluded-face-detection-on-mafa","task":"Occluded Face Detection","dataset":"MAFA","model":"AOFD","rank_in_archive_order":2,"of":2,"metrics":{"MAP":"77.3%"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}