{"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/invariant-teacher-and-equivariant-student-for","title":"Invariant Teacher and Equivariant Student for Unsupervised 3D Human Pose Estimation","arxiv_id":"2012.09398","date":"2020-12-17","proceeding":null,"authors":["Chenxin Xu","Siheng Chen","Maosen Li","Ya zhang"],"abstract":"We propose a novel method based on teacher-student learning framework for 3D human pose estimation without any 3D annotation or side information. To solve this unsupervised-learning problem, the teacher network adopts pose-dictionary-based modeling for regularization to estimate a physically plausible 3D pose. To handle the decomposition ambiguity in the teacher network, we propose a cycle-consistent architecture promoting a 3D rotation-invariant property to train the teacher network. To further improve the estimation accuracy, the student network adopts a novel graph convolution network for flexibility to directly estimate the 3D coordinates. Another cycle-consistent architecture promoting 3D rotation-equivariant property is adopted to exploit geometry consistency, together with knowledge distillation from the teacher network to improve the pose estimation performance. We conduct extensive experiments on Human3.6M and MPI-INF-3DHP. Our method reduces the 3D joint prediction error by 11.4% compared to state-of-the-art unsupervised methods and also outperforms many weakly-supervised methods that use side information on Human3.6M. Code will be available at https://github.com/sjtuxcx/ITES.","url_abs":"https://arxiv.org/abs/2012.09398v1","url_pdf":"https://arxiv.org/pdf/2012.09398v1.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":"invariant-teacher-and-equivariant-student-for","repo_url":"https://github.com/sjtuxcx/ITES","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"3d-human-pose-estimation","task_name":"3D Human Pose Estimation"},{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"},{"task_slug":"unsupervised-3d-human-pose-estimation","task_name":"Unsupervised 3D Human Pose Estimation"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"knowledge-distillation","method_name":"Knowledge Distillation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/unsupervised-3d-human-pose-estimation-on","task":"Unsupervised 3D Human Pose Estimation","dataset":"Human3.6M","model":"ITES-TS","rank_in_archive_order":4,"of":12,"metrics":{"MPJPE":"85.3","P-MPJPE":"59.8"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-3d-human-pose-estimation-on-mpi","task":"Unsupervised 3D Human Pose Estimation","dataset":"MPI-INF-3DHP","model":"ITES-TS","rank_in_archive_order":3,"of":4,"metrics":{"AUC":"35.2","PCK":"68.2"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2012.09398","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}