{"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/3d-human-pose-estimation-with-2d-marginal","title":"3D Human Pose Estimation with 2D Marginal Heatmaps","arxiv_id":"1806.01484","date":"2018-06-05","proceeding":null,"authors":["Aiden Nibali","Zhen He","Stuart Morgan","Luke Prendergast"],"abstract":"Automatically determining three-dimensional human pose from monocular RGB\nimage data is a challenging problem. The two-dimensional nature of the input\nresults in intrinsic ambiguities which make inferring depth particularly\ndifficult. Recently, researchers have demonstrated that the flexible\nstatistical modelling capabilities of deep neural networks are sufficient to\nmake such inferences with reasonable accuracy. However, many of these models\nuse coordinate output techniques which are memory-intensive, not\ndifferentiable, and/or do not spatially generalise well. We propose\nimprovements to 3D coordinate prediction which avoid the aforementioned\nundesirable traits by predicting 2D marginal heatmaps under an augmented\nsoft-argmax scheme. Our resulting model, MargiPose, produces visually coherent\nheatmaps whilst maintaining differentiability. We are also able to achieve\nstate-of-the-art accuracy on publicly available 3D human pose estimation data.","url_abs":"http://arxiv.org/abs/1806.01484v2","url_pdf":"http://arxiv.org/pdf/1806.01484v2.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":"3d-human-pose-estimation-with-2d-marginal","repo_url":"https://github.com/anibali/margipose","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"3d-human-pose-estimation","task_name":"3D Human Pose Estimation"},{"task_slug":"pose-estimation","task_name":"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":"MargiPose (multi-crop)","rank_in_archive_order":48,"of":108,"metrics":{"AUC":"47","MPJPE":"91.3","PCK":"85.4"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}