{"url":"/sota/multi-hypotheses-3d-human-pose-estimation-on","task":{"name":"Multi-Hypotheses 3D Human Pose Estimation","url":"/task/multi-hypotheses-3d-human-pose-estimation","note":null},"dataset":{"name":"Human3.6M","url":"/dataset/human3-6m"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":null,"description_from":null,"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 (mm)","Average PMPJPE (mm)","Average MPJPE (mm) for occluded Joints","Expected Calibration Error","Using 2D ground-truth joints"],"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 (mm)":null,"Average PMPJPE (mm)":null,"Average MPJPE (mm) for occluded Joints":null,"Expected Calibration Error":"lower","Using 2D ground-truth joints":null}},"counts":{"rows":12,"rows_with_code":11,"rows_with_paper_page":12,"rows_dated":12,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"DDHPose (H=20, W=10, J-Best)","metrics":{"Average MPJPE (mm)":"33.62","Average PMPJPE (mm)":"26.48"},"uses_additional_data":false,"paper_date":"2024-03-07","paper":"/paper/disentangled-diffusion-based-3d-human-pose","paper_url":"https://arxiv.org/abs/2403.04444v1","paper_title":"Disentangled Diffusion-Based 3D Human Pose Estimation with Hierarchical Spatial and Temporal Denoiser","code":"https://github.com/Andyen512/DDHPose","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"GFPose (HPJ2D-010, S=200)","metrics":{"Average MPJPE (mm)":"35.1"},"uses_additional_data":false,"paper_date":"2022-12-16","paper":"/paper/gfpose-learning-3d-human-pose-prior-with","paper_url":"https://arxiv.org/abs/2212.08641v1","paper_title":"GFPose: Learning 3D Human Pose Prior with Gradient Fields","code":"https://github.com/Embracing/GFPose","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"D3DP","metrics":{"Average MPJPE (mm)":"35.4","Using 2D ground-truth joints":"No"},"uses_additional_data":false,"paper_date":"2023-03-21","paper":"/paper/diffusion-based-3d-human-pose-estimation-with","paper_url":"https://arxiv.org/abs/2303.11579v2","paper_title":"Diffusion-Based 3D Human Pose Estimation with Multi-Hypothesis Aggregation","code":"https://github.com/patrick-swk/d3dp","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":4,"model":"GFPose (HPJ2D-000, S=200)","metrics":{"Average MPJPE (mm)":"35.6","Average PMPJPE (mm)":"30.5","Using 2D ground-truth joints":"16.9"},"uses_additional_data":false,"paper_date":"2022-12-16","paper":"/paper/gfpose-learning-3d-human-pose-prior-with","paper_url":"https://arxiv.org/abs/2212.08641v1","paper_title":"GFPose: Learning 3D Human Pose Prior with Gradient Fields","code":"https://github.com/Embracing/GFPose","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":5,"model":"DDHPose (H=20, W=10, P-Best)","metrics":{"Average MPJPE (mm)":"39.0","Average PMPJPE (mm)":"31.2"},"uses_additional_data":false,"paper_date":"2024-03-07","paper":"/paper/disentangled-diffusion-based-3d-human-pose","paper_url":"https://arxiv.org/abs/2403.04444v1","paper_title":"Disentangled Diffusion-Based 3D Human Pose Estimation with Hierarchical Spatial and Temporal Denoiser","code":"https://github.com/Andyen512/DDHPose","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"GraphMDN","metrics":{"Average MPJPE (mm)":"46.2","Average PMPJPE (mm)":"36.3"},"uses_additional_data":false,"paper_date":"2020-10-26","paper":"/paper/graphmdn-leveraging-graph-structure-and-deep","paper_url":"https://arxiv.org/abs/2010.13668v1","paper_title":"GraphMDN: Leveraging graph structure and deep learning to solve inverse problems","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"Sharma et al.","metrics":{"Average MPJPE (mm)":"46.8","Average PMPJPE (mm)":"37.3"},"uses_additional_data":false,"paper_date":"2019-04-02","paper":"/paper/monocular-3d-human-pose-estimation-by-1","paper_url":"https://arxiv.org/abs/1904.01324v2","paper_title":"Monocular 3D Human Pose Estimation by Generation and Ordinal Ranking","code":"https://github.com/ssfootball04/generative_pose","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":3,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"cGNF xlarge w Lsample","metrics":{"Average MPJPE (mm)":"48.5"},"uses_additional_data":false,"paper_date":"2022-10-20","paper":"/paper/multi-hypothesis-3d-human-pose-estimation","paper_url":"https://arxiv.org/abs/2210.11179v1","paper_title":"Multi-hypothesis 3D human pose estimation metrics favor miscalibrated distributions","code":"https://github.com/sinzlab/cgnf","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"MDN","metrics":{"Average MPJPE (mm)":"52.7","Average PMPJPE (mm)":"42.6"},"uses_additional_data":false,"paper_date":"2019-04-11","paper":"/paper/generating-multiple-hypotheses-for-3d-human","paper_url":"http://arxiv.org/abs/1904.05547v1","paper_title":"Generating Multiple Hypotheses for 3D Human Pose Estimation with Mixture Density Network","code":"https://github.com/chaneyddtt/Generating-Multiple-Hypotheses-for-3D-Human-Pose-Estimation-with-Mixture-Density-Network","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":10,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"cGNF w Lsample","metrics":{"Average MPJPE (mm)":"53","Average MPJPE (mm) for occluded Joints":"41.8","Expected Calibration Error":"0.08"},"uses_additional_data":false,"paper_date":"2022-10-20","paper":"/paper/multi-hypothesis-3d-human-pose-estimation","paper_url":"https://arxiv.org/abs/2210.11179v1","paper_title":"Multi-hypothesis 3D human pose estimation metrics favor miscalibrated distributions","code":"https://github.com/sinzlab/cgnf","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"Li et al.","metrics":{"Average MPJPE (mm)":"73.9","Average PMPJPE (mm)":"44.3"},"uses_additional_data":false,"paper_date":"2020-08-13","paper":"/paper/weakly-supervised-generative-network-for","paper_url":"https://arxiv.org/abs/2008.05770v1","paper_title":"Weakly Supervised Generative Network for Multiple 3D Human Pose Hypotheses","code":"https://github.com/chaneyddtt/weakly-supervised-3d-pose-generator","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":11,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"MHEntropy","metrics":{"Average PMPJPE (mm)":"36.8"},"uses_additional_data":false,"paper_date":"2023-01-01","paper":"/paper/mhentropy-entropy-meets-multiple-hypotheses","paper_url":"http://openaccess.thecvf.com//content/ICCV2023/html/Chen_MHEntropy_Entropy_Meets_Multiple_Hypotheses_for_Pose_and_Shape_Recovery_ICCV_2023_paper.html","paper_title":"MHEntropy: Entropy Meets Multiple Hypotheses for Pose and Shape Recovery","code":"https://github.com/GloryyrolG/MHEntropy","n_code_links":1,"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":3,"distinct_papers_with_graph_line":5,"distinct_papers_with_any_sample_ran":2,"samples_over_distinct_papers":{"n_ran":2,"n_unverified":27,"n_samples":29,"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":3,"n_unverified":28,"n_samples":31,"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"}}}