{"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/face-alignment-assisted-by-head-pose","title":"Face Alignment Assisted by Head Pose Estimation","arxiv_id":"1507.03148","date":"2015-07-11","proceeding":null,"authors":["Heng Yang","Wenxuan Mou","Yichi Zhang","Ioannis Patras","Hatice Gunes","Peter Robinson"],"abstract":"In this paper we propose a supervised initialization scheme for cascaded face\nalignment based on explicit head pose estimation. We first investigate the\nfailure cases of most state of the art face alignment approaches and observe\nthat these failures often share one common global property, i.e. the head pose\nvariation is usually large. Inspired by this, we propose a deep convolutional\nnetwork model for reliable and accurate head pose estimation. Instead of using\na mean face shape, or randomly selected shapes for cascaded face alignment\ninitialisation, we propose two schemes for generating initialisation: the first\none relies on projecting a mean 3D face shape (represented by 3D facial\nlandmarks) onto 2D image under the estimated head pose; the second one searches\nnearest neighbour shapes from the training set according to head pose distance.\nBy doing so, the initialisation gets closer to the actual shape, which enhances\nthe possibility of convergence and in turn improves the face alignment\nperformance. We demonstrate the proposed method on the benchmark 300W dataset\nand show very competitive performance in both head pose estimation and face\nalignment.","url_abs":"http://arxiv.org/abs/1507.03148v2","url_pdf":"http://arxiv.org/pdf/1507.03148v2.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":"face-alignment-assisted-by-head-pose","repo_url":"https://github.com/pocheck-v2/Pocheck-V2-Beta","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"face-alignment","task_name":"Face Alignment"},{"task_slug":"head-pose-estimation","task_name":"Head Pose Estimation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1507.03148","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}