{"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/monocular-3d-human-pose-estimation-in-the","title":"Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision","arxiv_id":"1611.09813","date":"2016-11-29","proceeding":null,"authors":["Dushyant Mehta","Helge Rhodin","Dan Casas","Pascal Fua","Oleksandr Sotnychenko","Weipeng Xu","Christian Theobalt"],"abstract":"We propose a CNN-based approach for 3D human body pose estimation from single\nRGB images that addresses the issue of limited generalizability of models\ntrained solely on the starkly limited publicly available 3D pose data. Using\nonly the existing 3D pose data and 2D pose data, we show state-of-the-art\nperformance on established benchmarks through transfer of learned features,\nwhile also generalizing to in-the-wild scenes. We further introduce a new\ntraining set for human body pose estimation from monocular images of real\nhumans that has the ground truth captured with a multi-camera marker-less\nmotion capture system. It complements existing corpora with greater diversity\nin pose, human appearance, clothing, occlusion, and viewpoints, and enables an\nincreased scope of augmentation. We also contribute a new benchmark that covers\noutdoor and indoor scenes, and demonstrate that our 3D pose dataset shows\nbetter in-the-wild performance than existing annotated data, which is further\nimproved in conjunction with transfer learning from 2D pose data. All in all,\nwe argue that the use of transfer learning of representations in tandem with\nalgorithmic and data contributions is crucial for general 3D body pose\nestimation.","url_abs":"http://arxiv.org/abs/1611.09813v5","url_pdf":"http://arxiv.org/pdf/1611.09813v5.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":[],"tasks":[{"task_slug":"3d-human-pose-estimation","task_name":"3D Human Pose Estimation"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"monocular-3d-human-pose-estimation","task_name":"Monocular 3D Human Pose Estimation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"methods":[],"datasets_introduced":[{"slug":"mpi-inf-3dhp","name":"MPI-INF-3DHP","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-human-pose-estimation-on-mpi-inf-3dhp","task":"3D Human Pose Estimation","dataset":"MPI-INF-3DHP","model":"Mehta","rank_in_archive_order":79,"of":108,"metrics":{"AUC":"39.3","MPJPE":"117.6","PCK":"75.7"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-mpi-inf-3dhp","task":"3D Human Pose Estimation","dataset":"MPI-INF-3DHP","model":"Mehta","rank_in_archive_order":97,"of":108,"metrics":{"AUC":"40.8","PCK":"64.7"},"uses_additional_data":false},{"leaderboard":"/sota/pose-estimation-on-leeds-sports-poses","task":"Pose Estimation","dataset":"Leeds Sports Poses","model":"Mehta","rank_in_archive_order":17,"of":18,"metrics":{"PCK":"75.7"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1611.09813","atlas_url":"https://app.syntology.ai/?focus=1611.09813","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}