{"url":"/sota/3d-hand-pose-estimation-on-freihand","task":{"name":"3D Hand Pose Estimation","url":"/task/3d-hand-pose-estimation","note":null},"dataset":{"name":"FreiHAND","url":"/dataset/freihand"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"Image: [Zimmerman et l](https://arxiv.xsrg/pdf/1705.01389v3.pdf)","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":["PA-MPJPE","PA-MPVPE","PA-F@5mm","PA-F@15mm"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"PA-MPJPE":null,"PA-MPVPE":null,"PA-F@5mm":null,"PA-F@15mm":null}},"counts":{"rows":33,"rows_with_code":23,"rows_with_paper_page":32,"rows_dated":32,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"ExtPose","metrics":{"PA-F@15mm":"0.993","PA-F@5mm":"0.823","PA-MPJPE":"4.9","PA-MPVPE":"5.1"},"uses_additional_data":false,"paper_date":"2025-06-18","paper":"/paper/extpose-robust-and-coherent-pose-estimation","paper_url":"https://openreview.net/forum?id=hm9FNEZZ6z","paper_title":"ExtPose: Robust and Coherent Pose Estimation by Extending ViTs","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":2,"model":"HandOS","metrics":{"PA-F@15mm":"0.991","PA-F@5mm":"0.812","PA-MPJPE":"5.0","PA-MPVPE":"5.3"},"uses_additional_data":false,"paper_date":"2024-12-02","paper":"/paper/handos-3d-hand-reconstruction-in-one-stage","paper_url":"https://arxiv.org/abs/2412.01537v2","paper_title":"HandOS: 3D Hand Reconstruction in One Stage","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":3,"model":"WiLoR","metrics":{"PA-F@15mm":"0.993","PA-F@5mm":"0.825","PA-MPJPE":"5.5","PA-MPVPE":"5.1"},"uses_additional_data":false,"paper_date":"2024-09-18","paper":"/paper/wilor-end-to-end-3d-hand-localization-and","paper_url":"https://arxiv.org/abs/2409.12259v2","paper_title":"WiLoR: End-to-end 3D Hand Localization and Reconstruction in-the-wild","code":"https://github.com/rolpotamias/WiLoR","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"MaskHand","metrics":{"PA-F@15mm":"0.991","PA-F@5mm":"0.801","PA-MPJPE":"5.5","PA-MPVPE":"5.4"},"uses_additional_data":false,"paper_date":"2024-12-18","paper":"/paper/mmhmr-generative-masked-modeling-for-hand","paper_url":"https://arxiv.org/abs/2412.13393v1","paper_title":"MMHMR: Generative Masked Modeling for Hand Mesh Recovery","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"Hamba","metrics":{"PA-F@15mm":"0.992","PA-F@5mm":"0.806","PA-MPJPE":"5.7","PA-MPVPE":"5.3"},"uses_additional_data":false,"paper_date":"2024-07-12","paper":"/paper/hamba-single-view-3d-hand-reconstruction-with","paper_url":"https://arxiv.org/abs/2407.09646v2","paper_title":"Hamba: Single-view 3D Hand Reconstruction with Graph-guided Bi-Scanning Mamba","code":"https://github.com/humansensinglab/Hamba","n_code_links":1,"syntology":{"n_ran":7,"n_unverified":4,"n_samples":11,"n_pointer_only_licence":11}},{"rank_in_archive_order":6,"model":"MobRecon","metrics":{"PA-F@15mm":"0.986","PA-F@5mm":"0.784","PA-MPJPE":"5.7","PA-MPVPE":"5.8"},"uses_additional_data":false,"paper_date":"2021-12-06","paper":"/paper/mobrecon-mobile-friendly-hand-mesh","paper_url":"https://arxiv.org/abs/2112.02753v2","paper_title":"MobRecon: Mobile-Friendly Hand Mesh Reconstruction from Monocular Image","code":"https://github.com/SeanChenxy/HandMesh","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":7,"n_samples":9,"n_pointer_only_licence":0}},{"rank_in_archive_order":7,"model":"Zhou et al.","metrics":{"PA-F@15mm":"0.986","PA-F@5mm":"0.772","PA-MPJPE":"5.7","PA-MPVPE":"6.0"},"uses_additional_data":false,"paper_date":"2024-03-04","paper":"/paper/a-simple-baseline-for-efficient-hand-mesh","paper_url":"https://arxiv.org/abs/2403.01813v1","paper_title":"A Simple Baseline for Efficient Hand Mesh Reconstruction","code":"https://github.com/patiencefromzhou/simplehand","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":14,"n_samples":15,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"HHMR","metrics":{"PA-MPJPE":"5.8","PA-MPVPE":"5.8"},"uses_additional_data":false,"paper_date":"2024-06-03","paper":"/paper/hhmr-holistic-hand-mesh-recovery-by-enhancing","paper_url":"https://arxiv.org/abs/2406.01334v1","paper_title":"HHMR: Holistic Hand Mesh Recovery by Enhancing the Multimodal Controllability of Graph Diffusion Models","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":9,"model":"MeshGraphormer","metrics":{"PA-F@15mm":"0.986","PA-F@5mm":"0.764","PA-MPJPE":"5.9","PA-MPVPE":"6.0"},"uses_additional_data":false,"paper_date":"2021-04-01","paper":"/paper/mesh-graphormer","paper_url":"https://arxiv.org/abs/2104.00272v2","paper_title":"Mesh Graphormer","code":"https://github.com/microsoft/meshgraphormer","n_code_links":3,"syntology":null},{"rank_in_archive_order":10,"model":"HaMeR","metrics":{"PA-F@15mm":"0.990","PA-F@5mm":"0.785","PA-MPJPE":"6.0","PA-MPVPE":"5.7"},"uses_additional_data":false,"paper_date":"2023-12-08","paper":"/paper/reconstructing-hands-in-3d-with-transformers","paper_url":"https://arxiv.org/abs/2312.05251v1","paper_title":"Reconstructing Hands in 3D with Transformers","code":"https://github.com/geopavlakos/hamer","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":6,"n_samples":6,"n_pointer_only_licence":0}},{"rank_in_archive_order":11,"model":"PointHMR","metrics":{"PA-F@15mm":"0.984","PA-F@5mm":"0.720","PA-MPJPE":"6.1","PA-MPVPE":"6.6"},"uses_additional_data":false,"paper_date":"2023-04-19","paper":"/paper/sampling-is-matter-point-guided-3d-human-mesh-1","paper_url":"https://arxiv.org/abs/2304.09502v1","paper_title":"Sampling is Matter: Point-guided 3D Human Mesh Reconstruction","code":"https://github.com/DCVL-3D/PointHMR_release","n_code_links":1,"syntology":{"n_ran":5,"n_unverified":3,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":12,"model":"AMVUR","metrics":{"PA-F@15mm":"0.987","PA-F@5mm":"0.767","PA-MPJPE":"6.2","PA-MPVPE":"6.1"},"uses_additional_data":false,"paper_date":"2023-04-27","paper":"/paper/a-probabilistic-attention-model-with","paper_url":"https://arxiv.org/abs/2304.14299v1","paper_title":"A Probabilistic Attention Model with Occlusion-aware Texture Regression for 3D Hand Reconstruction from a Single RGB Image","code":"https://github.com/zhehengjianglancaster/amvur","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"Deformer","metrics":{"PA-F@15mm":"0.984","PA-F@5mm":"0.743","PA-MPJPE":"6.2","PA-MPVPE":"6.4"},"uses_additional_data":false,"paper_date":"2023-01-01","paper":"/paper/deformable-mesh-transformer-for-3d-human-mesh","paper_url":"http://openaccess.thecvf.com//content/CVPR2023/html/Yoshiyasu_Deformable_Mesh_Transformer_for_3D_Human_Mesh_Recovery_CVPR_2023_paper.html","paper_title":"Deformable Mesh Transformer for 3D Human Mesh Recovery","code":"https://github.com/yusukey03012/deformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":14,"model":"SAR","metrics":{"PA-F@15mm":"0.981","PA-F@5mm":"0.724","PA-MPJPE":"6.3","PA-MPVPE":"6.5"},"uses_additional_data":false,"paper_date":null,"paper":null,"paper_url":null,"paper_title":"","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":15,"model":"METRO","metrics":{"PA-F@15mm":"0.984","PA-F@5mm":"0.731","PA-MPJPE":"6.5","PA-MPVPE":"6.3"},"uses_additional_data":false,"paper_date":"2020-12-17","paper":"/paper/end-to-end-human-pose-and-mesh-reconstruction","paper_url":"https://arxiv.org/abs/2012.09760v3","paper_title":"End-to-End Human Pose and Mesh Reconstruction with Transformers","code":"https://github.com/microsoft/MeshTransformer","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"FastMETRO","metrics":{"PA-F@15mm":"0.983","PA-F@5mm":"0.687","PA-MPJPE":"6.5","PA-MPVPE":"7.1"},"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":17,"model":"FastViT-MA36","metrics":{"PA-F@15mm":"0.981","PA-F@5mm":"0.722","PA-MPJPE":"6.6","PA-MPVPE":"6.7"},"uses_additional_data":false,"paper_date":"2023-03-24","paper":"/paper/fastvit-a-fast-hybrid-vision-transformer","paper_url":"https://arxiv.org/abs/2303.14189v2","paper_title":"FastViT: A Fast Hybrid Vision Transformer using Structural Reparameterization","code":"https://github.com/rwightman/pytorch-image-models","n_code_links":6,"syntology":{"n_ran":1,"n_unverified":4,"n_samples":5,"n_pointer_only_licence":5}},{"rank_in_archive_order":18,"model":"PeCLR","metrics":{"PA-MPJPE":"6.6"},"uses_additional_data":false,"paper_date":"2021-06-10","paper":"/paper/self-supervised-3d-hand-pose-estimation-from","paper_url":"https://arxiv.org/abs/2106.05953v5","paper_title":"PeCLR: Self-Supervised 3D Hand Pose Estimation from monocular RGB via Equivariant Contrastive Learning","code":"https://github.com/dahiyaaneesh/peclr","n_code_links":2,"syntology":null},{"rank_in_archive_order":19,"model":"Tang et al.","metrics":{"PA-F@15mm":"0.981","PA-F@5mm":"0.724","PA-MPJPE":"6.7","PA-MPVPE":"6.7"},"uses_additional_data":false,"paper_date":"2021-09-03","paper":"/paper/towards-accurate-alignment-in-real-time-3d","paper_url":"https://arxiv.org/abs/2109.01723v1","paper_title":"Towards Accurate Alignment in Real-time 3D Hand-Mesh Reconstruction","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":20,"model":"I2UV-HandNet","metrics":{"PA-F@15mm":"0.977","PA-F@5mm":"0.707","PA-MPJPE":"6.7","PA-MPVPE":"6.9"},"uses_additional_data":false,"paper_date":"2021-02-07","paper":"/paper/sr-affine-high-quality-3d-hand-model","paper_url":"https://arxiv.org/abs/2102.03725v2","paper_title":"I2UV-HandNet: Image-to-UV Prediction Network for Accurate and High-fidelity 3D Hand Mesh Modeling","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":21,"model":"CMR","metrics":{"PA-F@15mm":"0.977","PA-F@5mm":"0.715","PA-MPJPE":"6.9","PA-MPVPE":"7.0"},"uses_additional_data":false,"paper_date":"2021-03-04","paper":"/paper/camera-space-hand-mesh-recovery-via-semantic","paper_url":"https://arxiv.org/abs/2103.02845v2","paper_title":"Camera-Space Hand Mesh Recovery via Semantic Aggregation and Adaptive 2D-1D Registration","code":"https://github.com/SeanChenxy/HandMesh","n_code_links":1,"syntology":null},{"rank_in_archive_order":22,"model":"HIU-DMTL","metrics":{"PA-F@15mm":"0.974","PA-F@5mm":"0.699","PA-MPJPE":"7.1","PA-MPVPE":"7.3"},"uses_additional_data":false,"paper_date":"2021-07-24","paper":"/paper/hand-image-understanding-via-deep-multi-task","paper_url":"https://arxiv.org/abs/2107.11646v2","paper_title":"Hand Image Understanding via Deep Multi-Task Learning","code":"https://github.com/MandyMo/HIU-DMTL","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":1,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":23,"model":"I2L-MeshNet","metrics":{"PA-F@15mm":"0.973","PA-F@5mm":"0.681","PA-MPJPE":"7.4","PA-MPVPE":"7.6"},"uses_additional_data":false,"paper_date":"2020-08-09","paper":"/paper/i2l-meshnet-image-to-lixel-prediction-network-1","paper_url":"https://arxiv.org/abs/2008.03713v2","paper_title":"I2L-MeshNet: Image-to-Lixel Prediction Network for Accurate 3D Human Pose and Mesh Estimation from a Single RGB Image","code":"https://github.com/mks0601/I2L-MeshNet_RELEASE","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":1,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":24,"model":"Hand4Whole","metrics":{"PA-F@15mm":"0.971","PA-F@5mm":"0.664","PA-MPJPE":"7.7","PA-MPVPE":"7.7"},"uses_additional_data":false,"paper_date":"2020-11-23","paper":"/paper/pose2pose-3d-positional-pose-guided-3d","paper_url":"https://arxiv.org/abs/2011.11534v4","paper_title":"Accurate 3D Hand Pose Estimation for Whole-Body 3D Human Mesh Estimation","code":"https://github.com/mks0601/Hand4Whole_RELEASE","n_code_links":1,"syntology":null},{"rank_in_archive_order":25,"model":"Pose2Mesh","metrics":{"PA-F@15mm":"0.969","PA-F@5mm":"0.674","PA-MPJPE":"7.7","PA-MPVPE":"7.8"},"uses_additional_data":false,"paper_date":"2020-08-20","paper":"/paper/pose2mesh-graph-convolutional-network-for-3d","paper_url":"https://arxiv.org/abs/2008.09047v3","paper_title":"Pose2Mesh: Graph Convolutional Network for 3D Human Pose and Mesh Recovery from a 2D Human Pose","code":"https://github.com/hongsukchoi/Pose2Mesh_RELEASE","n_code_links":2,"syntology":null},{"rank_in_archive_order":26,"model":"YoutubeHand","metrics":{"PA-F@15mm":"0.966","PA-F@5mm":"0.614","PA-MPJPE":"8.4","PA-MPVPE":"8.6"},"uses_additional_data":false,"paper_date":"2020-04-04","paper":"/paper/weakly-supervised-mesh-convolutional-hand","paper_url":"https://arxiv.org/abs/2004.01946v1","paper_title":"Weakly-Supervised Mesh-Convolutional Hand Reconstruction in the Wild","code":"https://github.com/674106399/JointBoneLoss","n_code_links":4,"syntology":{"n_ran":6,"n_unverified":14,"n_samples":20,"n_pointer_only_licence":0}},{"rank_in_archive_order":27,"model":"MANO CNN","metrics":{"PA-F@15mm":"0.934","PA-F@5mm":"0.516","PA-MPJPE":"11.0","PA-MPVPE":"10.9"},"uses_additional_data":false,"paper_date":"2019-09-10","paper":"/paper/freihand-a-dataset-for-markerless-capture-of","paper_url":"https://arxiv.org/abs/1909.04349v3","paper_title":"FreiHAND: A Dataset for Markerless Capture of Hand Pose and Shape from Single RGB Images","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":28,"model":"PIXIE hand expert","metrics":{"PA-F@15mm":"0.919","PA-F@5mm":"0.468","PA-MPJPE":"12","PA-MPVPE":"12.1"},"uses_additional_data":false,"paper_date":"2021-05-11","paper":"/paper/collaborative-regression-of-expressive-bodies","paper_url":"https://arxiv.org/abs/2105.05301v2","paper_title":"Collaborative Regression of Expressive Bodies using Moderation","code":"https://github.com/YadiraF/PIXIE","n_code_links":1,"syntology":null},{"rank_in_archive_order":29,"model":"ExPose (hand sub-network h)","metrics":{"PA-F@15mm":"0.918","PA-F@5mm":"0.484","PA-MPJPE":"12.2","PA-MPVPE":"11.8"},"uses_additional_data":false,"paper_date":"2020-08-20","paper":"/paper/monocular-expressive-body-regression-through","paper_url":"https://arxiv.org/abs/2008.09062v1","paper_title":"Monocular Expressive Body Regression through Body-Driven Attention","code":"https://github.com/vchoutas/expose","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":30,"model":"DetNet","metrics":{"PA-MPJPE":"15.7"},"uses_additional_data":false,"paper_date":"2020-12-11","paper":"/paper/monocular-real-time-full-body-capture-with","paper_url":"https://arxiv.org/abs/2012.06087v2","paper_title":"Monocular Real-time Full Body Capture with Inter-part Correlations","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":31,"model":"Zimmermann et al.","metrics":{"PA-F@15mm":"0.935","PA-F@5mm":"0.529","PA-MPVPE":"10.7"},"uses_additional_data":false,"paper_date":"2019-09-10","paper":"/paper/freihand-a-dataset-for-markerless-capture-of","paper_url":"https://arxiv.org/abs/1909.04349v3","paper_title":"FreiHAND: A Dataset for Markerless Capture of Hand Pose and Shape from Single RGB Images","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":32,"model":"Boukhayma et al.","metrics":{"PA-F@15mm":"0.898","PA-F@5mm":"0.435","PA-MPVPE":"13.0"},"uses_additional_data":false,"paper_date":"2019-02-09","paper":"/paper/3d-hand-shape-and-pose-from-images-in-the","paper_url":"http://arxiv.org/abs/1902.03451v1","paper_title":"3D Hand Shape and Pose from Images in the Wild","code":"https://github.com/yihui-he/epipolar-transformers","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":33,"model":"Hasson et al.","metrics":{"PA-F@15mm":"0.908","PA-F@5mm":"0.436","PA-MPVPE":"13.2"},"uses_additional_data":false,"paper_date":"2019-04-11","paper":"/paper/learning-joint-reconstruction-of-hands-and","paper_url":"http://arxiv.org/abs/1904.05767v1","paper_title":"Learning joint reconstruction of hands and manipulated objects","code":"https://github.com/hassony2/manopth","n_code_links":3,"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":12,"rows_with_any_sample_ran":11,"distinct_papers_with_graph_line":12,"distinct_papers_with_any_sample_ran":11,"samples_over_distinct_papers":{"n_ran":43,"n_unverified":58,"n_samples":101,"n_pointer_only_licence":20,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":43,"n_unverified":58,"n_samples":101,"n_pointer_only_licence":20,"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"}}}