{"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/compositional-human-pose-regression","title":"Compositional Human Pose Regression","arxiv_id":"1704.00159","date":"2017-04-01","proceeding":"ICCV 2017 10","authors":["Xiao Sun","Jiaxiang Shang","Shuang Liang","Yichen Wei"],"abstract":"Regression based methods are not performing as well as detection based\nmethods for human pose estimation. A central problem is that the structural\ninformation in the pose is not well exploited in the previous regression\nmethods. In this work, we propose a structure-aware regression approach. It\nadopts a reparameterized pose representation using bones instead of joints. It\nexploits the joint connection structure to define a compositional loss function\nthat encodes the long range interactions in the pose. It is simple, effective,\nand general for both 2D and 3D pose estimation in a unified setting.\nComprehensive evaluation validates the effectiveness of our approach. It\nsignificantly advances the state-of-the-art on Human3.6M and is competitive\nwith state-of-the-art results on MPII.","url_abs":"http://arxiv.org/abs/1704.00159v3","url_pdf":"http://arxiv.org/pdf/1704.00159v3.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":"compositional-human-pose-regression","repo_url":"https://github.com/anibali/h36m-fetch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"3d-human-pose-estimation","task_name":"3D Human Pose Estimation"},{"task_slug":"3d-pose-estimation","task_name":"3D Pose Estimation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/pose-estimation-on-mpii-human-pose","task":"Pose Estimation","dataset":"MPII Human Pose","model":"CHPR","rank_in_archive_order":36,"of":46,"metrics":{"PCKh-0.5":"86.4"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.00159","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}