{"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/h3wb-human3-6m-3d-wholebody-dataset-and","title":"H3WB: Human3.6M 3D WholeBody Dataset and Benchmark","arxiv_id":"2211.15692","date":"2022-11-28","proceeding":"ICCV 2023 1","authors":["Yue Zhu","Nermin Samet","David Picard"],"abstract":"We present a benchmark for 3D human whole-body pose estimation, which involves identifying accurate 3D keypoints on the entire human body, including face, hands, body, and feet. Currently, the lack of a fully annotated and accurate 3D whole-body dataset results in deep networks being trained separately on specific body parts, which are combined during inference. Or they rely on pseudo-groundtruth provided by parametric body models which are not as accurate as detection based methods. To overcome these issues, we introduce the Human3.6M 3D WholeBody (H3WB) dataset, which provides whole-body annotations for the Human3.6M dataset using the COCO Wholebody layout. H3WB comprises 133 whole-body keypoint annotations on 100K images, made possible by our new multi-view pipeline. We also propose three tasks: i) 3D whole-body pose lifting from 2D complete whole-body pose, ii) 3D whole-body pose lifting from 2D incomplete whole-body pose, and iii) 3D whole-body pose estimation from a single RGB image. Additionally, we report several baselines from popular methods for these tasks. Furthermore, we also provide automated 3D whole-body annotations of TotalCapture and experimentally show that when used with H3WB it helps to improve the performance. Code and dataset is available at https://github.com/wholebody3d/wholebody3d","url_abs":"https://arxiv.org/abs/2211.15692v2","url_pdf":"https://arxiv.org/pdf/2211.15692v2.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":"h3wb-human3-6m-3d-wholebody-dataset-and","repo_url":"https://github.com/wholebody3d/wholebody3d","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"3d-facial-landmark-localization","task_name":"3D Facial Landmark Localization"},{"task_slug":"3d-hand-pose-estimation","task_name":"3D Hand Pose Estimation"},{"task_slug":"3d-human-pose-estimation","task_name":"3D Human Pose Estimation"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"}],"methods":[],"datasets_introduced":[{"slug":"h3wb","name":"H3WB","full_name":"Human 3.6M 3D WholeBody"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"Large SimpleBaseline","rank_in_archive_order":2,"of":15,"metrics":{"Average MPJPE (mm)":"14.6"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"Jointformer","rank_in_archive_order":4,"of":15,"metrics":{"Average MPJPE (mm)":"17.8"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"CanonPose + 3D supervision","rank_in_archive_order":5,"of":15,"metrics":{"Average MPJPE (mm)":"17.9"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"Large SimpleBaseline","rank_in_archive_order":6,"of":15,"metrics":{"Average MPJPE (mm)":"19.8"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"Jointformer","rank_in_archive_order":7,"of":15,"metrics":{"Average MPJPE (mm)":"19.8"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"CPN + Jointformer","rank_in_archive_order":8,"of":15,"metrics":{"Average MPJPE (mm)":" 20.7"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"CanonPose + 3D supervision","rank_in_archive_order":9,"of":15,"metrics":{"Average MPJPE (mm)":"22.2"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"CanonPose","rank_in_archive_order":10,"of":15,"metrics":{"Average MPJPE (mm)":"24.6"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"SimpleBaseline","rank_in_archive_order":11,"of":15,"metrics":{"Average MPJPE (mm)":"24.6"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"Resnet50","rank_in_archive_order":12,"of":15,"metrics":{"Average MPJPE (mm)":" 26.3"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"CanonPose","rank_in_archive_order":13,"of":15,"metrics":{"Average MPJPE (mm)":"31.9"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"SHN + SimpleBaseline","rank_in_archive_order":14,"of":15,"metrics":{"Average MPJPE (mm)":"32.5"},"uses_additional_data":false},{"leaderboard":"/sota/3d-facial-landmark-localization-on-h3wb","task":"3D Facial Landmark Localization","dataset":"H3WB","model":"SimpleBaseline","rank_in_archive_order":15,"of":15,"metrics":{"Average MPJPE (mm)":"34.0"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"Large SimpleBaseline","rank_in_archive_order":3,"of":15,"metrics":{"Average MPJPE (mm)":"31.7"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"CanonPose + 3D supervision","rank_in_archive_order":4,"of":15,"metrics":{"Average MPJPE (mm)":"38.3"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"SimpleBaseline","rank_in_archive_order":5,"of":15,"metrics":{"Average MPJPE (mm)":"42.5"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"Jointformer","rank_in_archive_order":6,"of":15,"metrics":{"Average MPJPE (mm)":"43.7"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"Large SimpleBaseline","rank_in_archive_order":7,"of":15,"metrics":{"Average MPJPE (mm)":"44.8"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"CanonPose + 3D supervision","rank_in_archive_order":8,"of":15,"metrics":{"Average MPJPE (mm)":"47.4"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"CanonPose","rank_in_archive_order":9,"of":15,"metrics":{"Average MPJPE (mm)":"48.9"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"Jointformer","rank_in_archive_order":10,"of":15,"metrics":{"Average MPJPE (mm)":"53.5"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"CanonPose","rank_in_archive_order":11,"of":15,"metrics":{"Average MPJPE (mm)":"56.2"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"CPN + Jointformer","rank_in_archive_order":12,"of":15,"metrics":{"Average MPJPE (mm)":"56.9"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"Resnet50","rank_in_archive_order":13,"of":15,"metrics":{"Average MPJPE (mm)":"63.1"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"SHN + SimpleBaseline","rank_in_archive_order":14,"of":15,"metrics":{"Average MPJPE (mm)":"64.3"},"uses_additional_data":false},{"leaderboard":"/sota/3d-hand-pose-estimation-on-h3wb","task":"3D Hand Pose Estimation","dataset":"H3WB","model":"SimpleBaseline","rank_in_archive_order":15,"of":15,"metrics":{"Average MPJPE (mm)":"83.4"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"Jointformer","rank_in_archive_order":5,"of":17,"metrics":{"MPJPE":"84.9"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"Jointformer","rank_in_archive_order":6,"of":17,"metrics":{"MPJPE":"103.0"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"Large SimpleBaseline","rank_in_archive_order":7,"of":17,"metrics":{"MPJPE":"112.6"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"CanonPose + 3D supervision","rank_in_archive_order":8,"of":17,"metrics":{"MPJPE":"117.5"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"SimpleBaseline","rank_in_archive_order":9,"of":17,"metrics":{"MPJPE":"125.7"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"Large SimpleBaseline","rank_in_archive_order":10,"of":17,"metrics":{"MPJPE":"131.6"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"CPN + Jointformer","rank_in_archive_order":11,"of":17,"metrics":{"MPJPE":"142.8"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"Resnet50","rank_in_archive_order":12,"of":17,"metrics":{"MPJPE":"151.6"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"CanonPose + 3D supervision","rank_in_archive_order":13,"of":17,"metrics":{"MPJPE":"155.9"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"SHN + SimpleBaseline","rank_in_archive_order":14,"of":17,"metrics":{"MPJPE":"189.6"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"CanonPose","rank_in_archive_order":15,"of":17,"metrics":{"MPJPE":"193.7"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"SimpleBaseline","rank_in_archive_order":16,"of":17,"metrics":{"MPJPE":"252.0"},"uses_additional_data":false},{"leaderboard":"/sota/3d-human-pose-estimation-on-h3wb","task":"3D Human Pose Estimation","dataset":"H3WB","model":"CanonPose","rank_in_archive_order":17,"of":17,"metrics":{"MPJPE":"264.4"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2211.15692","atlas_url":"https://app.syntology.ai/?focus=2211.15692","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}