{"url":"/sota/3d-human-pose-estimation-on-emdb","task":{"name":"3D Human Pose Estimation","url":"/task/3d-human-pose-estimation","note":null},"dataset":{"name":"EMDB","url":"/dataset/emdb"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"**3D Human Pose Estimation** is a computer vision task that involves estimating the 3D positions and orientations of body joints and bones from 2D images or videos. The goal is to reconstruct the 3D pose of a person in real-time, which can be used in a variety of applications, such as virtual reality, human-computer interaction, and motion analysis.","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["Average MPJPE-PA (mm)","Average MPJPE (mm)","Average MVE (mm)","Average MVE-PA (mm)","Average MPJAE (deg)","Average MPJAE-PA (deg)","Jitter (10m/s^3)"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Average MPJPE-PA (mm)":null,"Average MPJPE (mm)":null,"Average MVE (mm)":null,"Average MVE-PA (mm)":null,"Average MPJAE (deg)":null,"Average MPJAE-PA (deg)":null,"Jitter (10m/s^3)":null}},"counts":{"rows":13,"rows_with_code":10,"rows_with_paper_page":13,"rows_dated":13,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"TRAM","metrics":{"Average MPJPE (mm)":"74.4","Average MPJPE-PA (mm)":"45.7","Average MVE (mm)":"86.6"},"uses_additional_data":false,"paper_date":"2024-03-26","paper":"/paper/tram-global-trajectory-and-motion-of-3d","paper_url":"https://arxiv.org/abs/2403.17346v2","paper_title":"TRAM: Global Trajectory and Motion of 3D Humans from in-the-wild Videos","code":"https://github.com/yufu-wang/tram","n_code_links":1,"syntology":{"n_ran":13,"n_unverified":1,"n_samples":14,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"WHAM (ViT)","metrics":{"Average MPJPE (mm)":"79.7","Average MPJPE-PA (mm)":"50.4","Average MVE (mm)":"94.4"},"uses_additional_data":false,"paper_date":"2023-12-12","paper":"/paper/wham-reconstructing-world-grounded-humans","paper_url":"https://arxiv.org/abs/2312.07531v2","paper_title":"WHAM: Reconstructing World-grounded Humans with Accurate 3D Motion","code":"https://github.com/yohanshin/WHAM","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":9,"n_samples":12,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"GenHMR","metrics":{"Average MPJPE (mm)":"88.2","Average MPJPE-PA (mm)":"51.7","Average MVE (mm)":"99.5"},"uses_additional_data":false,"paper_date":"2024-12-19","paper":"/paper/genhmr-generative-human-mesh-recovery","paper_url":"https://arxiv.org/abs/2412.14444v1","paper_title":"GenHMR: Generative Human Mesh Recovery","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":4,"model":"BioPose","metrics":{"Average MPJPE (mm)":"92.5","Average MPJPE-PA (mm)":"52.1","Average MVE (mm)":"98.9"},"uses_additional_data":false,"paper_date":"2025-01-14","paper":"/paper/biopose-biomechanically-accurate-3d-pose","paper_url":"https://arxiv.org/abs/2501.07800v1","paper_title":"BioPose: Biomechanically-accurate 3D Pose Estimation from Monocular Videos","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"HybrIK","metrics":{"Average MPJAE (deg)":"24.5174","Average MPJAE-PA (deg)":"23.0704","Average MPJPE (mm)":"103.037","Average MPJPE-PA (mm)":"65.5935","Average MVE (mm)":"122.193","Average MVE-PA (mm)":"80.3678","Jitter (10m/s^3)":"49.2068"},"uses_additional_data":false,"paper_date":"2020-11-30","paper":"/paper/hybrik-a-hybrid-analytical-neural-inverse","paper_url":"https://arxiv.org/abs/2011.14672v4","paper_title":"HybrIK: A Hybrid Analytical-Neural Inverse Kinematics Solution for 3D Human Pose and Shape Estimation","code":"https://github.com/jeffffffli/HybrIK","n_code_links":4,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":6,"model":"FastMETRO-L no SMPL Head","metrics":{"Average MPJPE (mm)":"108.107","Average MPJPE-PA (mm)":"66.794","Average MVE (mm)":"119.23","Average MVE-PA (mm)":"81.1847","Jitter (10m/s^3)":"185.933"},"uses_additional_data":false,"paper_date":"2022-07-27","paper":"/paper/cross-attention-of-disentangled-modalities","paper_url":"https://arxiv.org/abs/2207.13820v1","paper_title":"Cross-Attention of Disentangled Modalities for 3D Human Mesh Recovery with Transformers","code":"https://github.com/postech-ami/fastmetro","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":4,"n_samples":16,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"CLIFF","metrics":{"Average MPJAE (deg)":"23.0933","Average MPJAE-PA (deg)":"21.6265","Average MPJPE (mm)":"103.134","Average MPJPE-PA (mm)":"68.7969","Average MVE (mm)":"122.884","Average MVE-PA (mm)":"81.3275","Jitter (10m/s^3)":"55.4525"},"uses_additional_data":false,"paper_date":"2022-08-01","paper":"/paper/cliff-carrying-location-information-in-full","paper_url":"https://arxiv.org/abs/2208.00571v2","paper_title":"CLIFF: Carrying Location Information in Full Frames into Human Pose and Shape Estimation","code":"https://github.com/huawei-noah/noah-research/tree/master/CLIFF","n_code_links":6,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"GLAMR","metrics":{"Average MPJAE (deg)":"25.5058","Average MPJAE-PA (deg)":"23.5484","Average MPJPE (mm)":"107.797","Average MPJPE-PA (mm)":"71.0091","Average MVE (mm)":"128.237","Average MVE-PA (mm)":"85.4586","Jitter (10m/s^3)":"67.3522"},"uses_additional_data":false,"paper_date":"2021-12-02","paper":"/paper/glamr-global-occlusion-aware-human-mesh","paper_url":"https://arxiv.org/abs/2112.01524v2","paper_title":"GLAMR: Global Occlusion-Aware Human Mesh Recovery with Dynamic Cameras","code":"https://github.com/nvlabs/glamr","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"PARE","metrics":{"Average MPJAE (deg)":"24.673","Average MPJAE-PA (deg)":"22.3842","Average MPJPE (mm)":"113.887","Average MPJPE-PA (mm)":"72.203","Average MVE (mm)":"133.247","Average MVE-PA (mm)":"85.3788","Jitter (10m/s^3)":"75.1137"},"uses_additional_data":false,"paper_date":"2021-04-17","paper":"/paper/pare-part-attention-regressor-for-3d-human","paper_url":"https://arxiv.org/abs/2104.08527v2","paper_title":"PARE: Part Attention Regressor for 3D Human Body Estimation","code":"https://github.com/mkocabas/PARE","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"FastMETRO-L","metrics":{"Average MPJAE (deg)":"25.07","Average MPJAE-PA (deg)":"22.9482","Average MPJPE (mm)":"115.036","Average MPJPE-PA (mm)":"72.6765","Average MVE (mm)":"133.566","Average MVE-PA (mm)":"86.0043","Jitter (10m/s^3)":"81.2959"},"uses_additional_data":false,"paper_date":"2022-07-27","paper":"/paper/cross-attention-of-disentangled-modalities","paper_url":"https://arxiv.org/abs/2207.13820v1","paper_title":"Cross-Attention of Disentangled Modalities for 3D Human Mesh Recovery with Transformers","code":"https://github.com/postech-ami/fastmetro","n_code_links":1,"syntology":{"n_ran":12,"n_unverified":4,"n_samples":16,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"ROMP","metrics":{"Average MPJAE (deg)":"26.5975","Average MPJAE-PA (deg)":"23.9901","Average MPJPE (mm)":"112.652","Average MPJPE-PA (mm)":"75.1869","Average MVE (mm)":"134.863","Average MVE-PA (mm)":"90.648","Jitter (10m/s^3)":"71.2556"},"uses_additional_data":false,"paper_date":"2020-08-27","paper":"/paper/centerhmr-a-bottom-up-single-shot-method-for","paper_url":"https://arxiv.org/abs/2008.12272v4","paper_title":"Monocular, One-stage, Regression of Multiple 3D People","code":"https://github.com/Arthur151/ROMP","n_code_links":2,"syntology":null},{"rank_in_archive_order":12,"model":"LGD","metrics":{"Average MPJAE (deg)":"25.1572","Average MPJAE-PA (deg)":"25.6425","Average MPJPE (mm)":"115.82","Average MPJPE-PA (mm)":"81.1389","Average MVE (mm)":"140.63","Average MVE-PA (mm)":"95.7255","Jitter (10m/s^3)":"72.9695"},"uses_additional_data":false,"paper_date":"2020-08-19","paper":"/paper/human-body-model-fitting-by-learned-gradient","paper_url":"https://arxiv.org/abs/2008.08474v1","paper_title":"Human Body Model Fitting by Learned Gradient Descent","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"PyMAF","metrics":{"Average MPJAE (deg)":"28.4555","Average MPJAE-PA (deg)":"25.7033","Average MPJPE (mm)":"131.065","Average MPJPE-PA (mm)":"82.8502","Average MVE (mm)":"159.956","Average MVE-PA (mm)":"98.1305","Jitter (10m/s^3)":"81.8447"},"uses_additional_data":false,"paper_date":"2021-03-30","paper":"/paper/3d-human-pose-and-shape-regression-with","paper_url":"https://arxiv.org/abs/2103.16507v4","paper_title":"PyMAF: 3D Human Pose and Shape Regression with Pyramidal Mesh Alignment Feedback Loop","code":"https://github.com/HongwenZhang/PyMAF","n_code_links":2,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,264 of the 9,581 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9581,"papers_checked":6264,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":3316},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":6,"rows_with_any_sample_ran":5,"distinct_papers_with_graph_line":5,"distinct_papers_with_any_sample_ran":4,"samples_over_distinct_papers":{"n_ran":29,"n_unverified":17,"n_samples":46,"n_pointer_only_licence":0,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":41,"n_unverified":21,"n_samples":62,"n_pointer_only_licence":0,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}