{"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/repnet-weakly-supervised-training-of-an","title":"RepNet: Weakly Supervised Training of an Adversarial Reprojection Network for 3D Human Pose Estimation","arxiv_id":"1902.09868","date":"2019-02-26","proceeding":"CVPR 2019 6","authors":["Bastian Wandt","Bodo Rosenhahn"],"abstract":"This paper addresses the problem of 3D human pose estimation from single\nimages. While for a long time human skeletons were parameterized and fitted to\nthe observation by satisfying a reprojection error, nowadays researchers\ndirectly use neural networks to infer the 3D pose from the observations.\nHowever, most of these approaches ignore the fact that a reprojection\nconstraint has to be satisfied and are sensitive to overfitting. We tackle the\noverfitting problem by ignoring 2D to 3D correspondences. This efficiently\navoids a simple memorization of the training data and allows for a weakly\nsupervised training. One part of the proposed reprojection network (RepNet)\nlearns a mapping from a distribution of 2D poses to a distribution of 3D poses\nusing an adversarial training approach. Another part of the network estimates\nthe camera. This allows for the definition of a network layer that performs the\nreprojection of the estimated 3D pose back to 2D which results in a\nreprojection loss function. Our experiments show that RepNet generalizes well\nto unknown data and outperforms state-of-the-art methods when applied to unseen\ndata. Moreover, our implementation runs in real-time on a standard desktop PC.","url_abs":"http://arxiv.org/abs/1902.09868v2","url_pdf":"http://arxiv.org/pdf/1902.09868v2.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":"repnet-weakly-supervised-training-of-an","repo_url":"https://github.com/bastianwandt/RepNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"3d-human-pose-estimation","task_name":"3D Human Pose Estimation"},{"task_slug":"memorization","task_name":"Memorization"},{"task_slug":"monocular-3d-human-pose-estimation","task_name":"Monocular 3D Human Pose Estimation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"},{"task_slug":"weakly-supervised-3d-human-pose-estimation","task_name":"Weakly-supervised 3D Human Pose Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-human-pose-estimation-on-mpi-inf-3dhp","task":"3D Human Pose Estimation","dataset":"MPI-INF-3DHP","model":"RepNet (H36M)","rank_in_archive_order":49,"of":108,"metrics":{"AUC":"54.8","MPJPE":"92.5","PCK":"81.8"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-mpi-inf-3dhp","task":"3D Human Pose Estimation","dataset":"MPI-INF-3DHP","model":"RepNet (3DHP)","rank_in_archive_order":65,"of":108,"metrics":{"AUC":"58.5","MPJPE":"97.8","PCK":"82.5"},"uses_additional_data":false},{"leaderboard":"/sota/monocular-3d-human-pose-estimation-on-human3","task":"Monocular 3D Human Pose Estimation","dataset":"Human3.6M","model":"RepNet","rank_in_archive_order":38,"of":52,"metrics":{"Average MPJPE (mm)":"89.9","Frames Needed":"1","Need Ground Truth 2D Pose":"No","Use Video Sequence":"No"},"uses_additional_data":false},{"leaderboard":"/sota/weakly-supervised-3d-human-pose-estimation-on","task":"Weakly-supervised 3D Human Pose Estimation","dataset":"Human3.6M","model":"RepNet","rank_in_archive_order":24,"of":33,"metrics":{"3D Annotations":"No","Average MPJPE (mm)":"89.9","Number of Frames Per View":"1","Number of Views":"1"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1902.09868","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}