{"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/look-at-boundary-a-boundary-aware-face","title":"Look at Boundary: A Boundary-Aware Face Alignment Algorithm","arxiv_id":"1805.10483","date":"2018-05-26","proceeding":"CVPR 2018 6","authors":["Wayne Wu","Chen Qian","Shuo Yang","Quan Wang","Yici Cai","Qiang Zhou"],"abstract":"We present a novel boundary-aware face alignment algorithm by utilising\nboundary lines as the geometric structure of a human face to help facial\nlandmark localisation. Unlike the conventional heatmap based method and\nregression based method, our approach derives face landmarks from boundary\nlines which remove the ambiguities in the landmark definition. Three questions\nare explored and answered by this work: 1. Why using boundary? 2. How to use\nboundary? 3. What is the relationship between boundary estimation and landmarks\nlocalisation? Our boundary- aware face alignment algorithm achieves 3.49% mean\nerror on 300-W Fullset, which outperforms state-of-the-art methods by a large\nmargin. Our method can also easily integrate information from other datasets.\nBy utilising boundary information of 300-W dataset, our method achieves 3.92%\nmean error with 0.39% failure rate on COFW dataset, and 1.25% mean error on\nAFLW-Full dataset. Moreover, we propose a new dataset WFLW to unify training\nand testing across different factors, including poses, expressions,\nilluminations, makeups, occlusions, and blurriness. Dataset and model will be\npublicly available at https://wywu.github.io/projects/LAB/LAB.html","url_abs":"http://arxiv.org/abs/1805.10483v1","url_pdf":"http://arxiv.org/pdf/1805.10483v1.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":"look-at-boundary-a-boundary-aware-face","repo_url":"https://github.com/wywu/LAB","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"look-at-boundary-a-boundary-aware-face","repo_url":"https://github.com/open-mmlab/mmpose","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"face-alignment","task_name":"Face Alignment"},{"task_slug":"facial-landmark-detection","task_name":"Facial Landmark Detection"}],"methods":[{"method_slug":"heatmap","method_name":"Heatmap"}],"datasets_introduced":[{"slug":"wflw","name":"WFLW","full_name":"Wider Facial Landmarks in the Wild"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/face-alignment-on-300w","task":"Face Alignment","dataset":"300W","model":"LAB","rank_in_archive_order":29,"of":48,"metrics":{"NME_inter-ocular (%, Challenge)":"5.19","NME_inter-ocular (%, Common)":"2.98","NME_inter-ocular (%, Full)":"3.49","NME_inter-pupil (%, Challenge)":"6.98","NME_inter-pupil (%, Common)":"3.42","NME_inter-pupil (%, Full)":"4.12"},"uses_additional_data":true},{"leaderboard":"/sota/face-alignment-on-aflw-19","task":"Face Alignment","dataset":"AFLW-19","model":"LAB (w/ B)","rank_in_archive_order":4,"of":23,"metrics":{"NME_diag (%, Frontal)":"1.14","NME_diag (%, Full)":"1.25"},"uses_additional_data":true},{"leaderboard":"/sota/face-alignment-on-aflw-19","task":"Face Alignment","dataset":"AFLW-19","model":"LAB (w/o B)","rank_in_archive_order":15,"of":23,"metrics":{"NME_diag (%, Frontal)":"1.62","NME_diag (%, Full)":"1.85"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-cofw","task":"Face Alignment","dataset":"COFW","model":"LAB (w/ B)","rank_in_archive_order":18,"of":28,"metrics":{"NME (inter-ocular)":"3.92%"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-cofw","task":"Face Alignment","dataset":"COFW","model":"LAB","rank_in_archive_order":21,"of":28,"metrics":{"NME (inter-ocular)":"5.58%"},"uses_additional_data":false},{"leaderboard":"/sota/face-alignment-on-wflw","task":"Face Alignment","dataset":"WFLW","model":"LAB","rank_in_archive_order":29,"of":36,"metrics":{"AUC@10 (inter-ocular)":"53.2","FR@10 (inter-ocular)":"7.56","NME (inter-ocular)":"5.27"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1805.10483","atlas_url":"https://app.syntology.ai/?focus=1805.10483","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}