{"url":"/sota/pose-estimation-on-leeds-sports-poses","task":{"name":"Pose Estimation","url":"/task/pose-estimation","note":null},"dataset":{"name":"Leeds Sports Poses","url":"/dataset/lsp"},"category":"Computer Vision","categories":["Computer Vision"],"category_note":null,"description":"**Pose Estimation** is a computer vision task where the goal is to detect the position and orientation of a person or an object. Usually, this is done by predicting the location of specific keypoints like hands, head, elbows, etc. in case of Human Pose Estimation.\r\n\r\nA common benchmark for this task is [MPII Human Pose](https://paperswithcode.com/sota/pose-estimation-on-mpii-human-pose)\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Real-time 2D Multi-Person Pose Estimation on CPU: Lightweight OpenPose](https://github.com/Daniil-Osokin/lightweight-human-pose-estimation.pytorch) )</span>","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":["PCK"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"PCK":null}},"counts":{"rows":18,"rows_with_code":14,"rows_with_paper_page":18,"rows_dated":18,"rows_using_additional_data":0},"rows":[{"rank_in_archive_order":1,"model":"OmniPose","metrics":{"PCK":"99.5%"},"uses_additional_data":false,"paper_date":"2021-03-18","paper":"/paper/omnipose-a-multi-scale-framework-for-multi","paper_url":"https://arxiv.org/abs/2103.10180v1","paper_title":"OmniPose: A Multi-Scale Framework for Multi-Person Pose Estimation","code":"https://github.com/bmartacho/OmniPose","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"Soft-gated Skip Connections","metrics":{"PCK":"94.8%"},"uses_additional_data":false,"paper_date":"2020-02-25","paper":"/paper/toward-fast-and-accurate-human-pose","paper_url":"https://arxiv.org/abs/2002.11098v1","paper_title":"Toward fast and accurate human pose estimation via soft-gated skip connections","code":"https://github.com/BB-Repos/BBpose","n_code_links":3,"syntology":null},{"rank_in_archive_order":3,"model":"UniPose","metrics":{"PCK":"94.5%"},"uses_additional_data":false,"paper_date":"2020-01-22","paper":"/paper/unipose-unified-human-pose-estimation-in","paper_url":"https://arxiv.org/abs/2001.08095v1","paper_title":"UniPose: Unified Human Pose Estimation in Single Images and Videos","code":"https://github.com/bmartacho/UniPose","n_code_links":2,"syntology":null},{"rank_in_archive_order":4,"model":"Residual Hourglass + ASR + AHO","metrics":{"PCK":"94.5%"},"uses_additional_data":false,"paper_date":"2018-05-24","paper":"/paper/jointly-optimize-data-augmentation-and","paper_url":"http://arxiv.org/abs/1805.09707v1","paper_title":"Jointly Optimize Data Augmentation and Network Training: Adversarial Data Augmentation in Human Pose Estimation","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":5,"model":"Chou et al. arXiv'17","metrics":{"PCK":"94%"},"uses_additional_data":false,"paper_date":"2017-07-08","paper":"/paper/self-adversarial-training-for-human-pose","paper_url":"http://arxiv.org/abs/1707.02439v2","paper_title":"Self Adversarial Training for Human Pose Estimation","code":"https://github.com/dongzhuoyao/jessiechouuu-adversarial-pose","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"Pyramid Residual Modules (PRMs)","metrics":{"PCK":"93.9%"},"uses_additional_data":false,"paper_date":"2017-08-03","paper":"/paper/learning-feature-pyramids-for-human-pose","paper_url":"http://arxiv.org/abs/1708.01101v1","paper_title":"Learning Feature Pyramids for Human Pose Estimation","code":"https://github.com/bearpaw/PyraNet","n_code_links":3,"syntology":null},{"rank_in_archive_order":7,"model":"Stacked hourglass + Inception-resnet","metrics":{"PCK":"93.9%"},"uses_additional_data":false,"paper_date":"2017-05-05","paper":"/paper/knowledge-guided-deep-fractal-neural-networks","paper_url":"http://arxiv.org/abs/1705.02407v2","paper_title":"Knowledge-Guided Deep Fractal Neural Networks for Human Pose Estimation","code":"https://github.com/Guanghan/GNet-pose","n_code_links":1,"syntology":null},{"rank_in_archive_order":8,"model":"Multi-Context Attention","metrics":{"PCK":"92.6%"},"uses_additional_data":false,"paper_date":"2017-02-24","paper":"/paper/multi-context-attention-for-human-pose","paper_url":"http://arxiv.org/abs/1702.07432v1","paper_title":"Multi-Context Attention for Human Pose Estimation","code":"https://github.com/wbenbihi/hourglasstensorlfow","n_code_links":2,"syntology":null},{"rank_in_archive_order":9,"model":"FPD","metrics":{"PCK":"90.8%"},"uses_additional_data":false,"paper_date":"2018-11-13","paper":"/paper/fast-human-pose-estimation","paper_url":"http://arxiv.org/abs/1811.05419v2","paper_title":"Fast Human Pose Estimation","code":"https://github.com/ilovepose/fast-human-pose-estimation.pytorch","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"Part heatmap regression (ResNet-152)","metrics":{"PCK":"90.7%"},"uses_additional_data":false,"paper_date":"2016-09-06","paper":"/paper/human-pose-estimation-via-convolutional-part","paper_url":"http://arxiv.org/abs/1609.01743v1","paper_title":"Human pose estimation via Convolutional Part Heatmap Regression","code":"https://github.com/1adrianb/human-pose-estimation","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"Convolutional Pose Machines","metrics":{"PCK":"90.5%"},"uses_additional_data":false,"paper_date":"2016-01-30","paper":"/paper/convolutional-pose-machines","paper_url":"http://arxiv.org/abs/1602.00134v4","paper_title":"Convolutional Pose Machines","code":"https://github.com/CMU-Perceptual-Computing-Lab/openpose","n_code_links":50,"syntology":{"n_ran":1,"n_unverified":3,"n_samples":4,"n_pointer_only_licence":3}},{"rank_in_archive_order":12,"model":"Soft-argmax + contextual information","metrics":{"PCK":"90.5%"},"uses_additional_data":false,"paper_date":"2017-10-06","paper":"/paper/human-pose-regression-by-combining-indirect","paper_url":"http://arxiv.org/abs/1710.02322v1","paper_title":"Human Pose Regression by Combining Indirect Part Detection and Contextual Information","code":"https://github.com/dluvizon/pose-regression","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":0}},{"rank_in_archive_order":13,"model":"ResNet-152 + intermediate supervision","metrics":{"PCK":"90.1%"},"uses_additional_data":false,"paper_date":"2016-05-10","paper":"/paper/deepercut-a-deeper-stronger-and-faster-multi","paper_url":"http://arxiv.org/abs/1605.03170v3","paper_title":"DeeperCut: A Deeper, Stronger, and Faster Multi-Person Pose Estimation Model","code":"https://github.com/eldar/pose-tensorflow","n_code_links":16,"syntology":null},{"rank_in_archive_order":14,"model":"DG-Net (T=4)","metrics":{"PCK":"87.5%"},"uses_additional_data":false,"paper_date":"2021-09-15","paper":"/paper/learning-dynamical-human-joint-affinity-for","paper_url":"https://arxiv.org/abs/2109.07353v1","paper_title":"Learning Dynamical Human-Joint Affinity for 3D Pose Estimation in Videos","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":15,"model":"Trajectory Space Factorization (F=25)","metrics":{"PCK":"83.6"},"uses_additional_data":false,"paper_date":"2019-08-22","paper":"/paper/trajectory-space-factorization-for-deep-video","paper_url":"https://arxiv.org/abs/1908.08289v1","paper_title":"Trajectory Space Factorization for Deep Video-Based 3D Human Pose Estimation","code":"https://github.com/jiahaoLjh/trajectory-pose-3d","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"VNect (ResNet 50)","metrics":{"PCK":"79.4"},"uses_additional_data":false,"paper_date":"2017-05-03","paper":"/paper/vnect-real-time-3d-human-pose-estimation-with","paper_url":"http://arxiv.org/abs/1705.01583v1","paper_title":"VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera","code":"https://github.com/XinArkh/VNect","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"Mehta","metrics":{"PCK":"75.7"},"uses_additional_data":false,"paper_date":"2016-11-29","paper":"/paper/monocular-3d-human-pose-estimation-in-the","paper_url":"http://arxiv.org/abs/1611.09813v5","paper_title":"Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":18,"model":"Chen&Yuille, NIPS'14","metrics":{"PCK":"73.4%"},"uses_additional_data":false,"paper_date":"2014-07-12","paper":"/paper/articulated-pose-estimation-by-a-graphical","paper_url":"http://arxiv.org/abs/1407.3399v2","paper_title":"Articulated Pose Estimation by a Graphical Model with Image Dependent Pairwise Relations","code":null,"n_code_links":0,"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":2,"rows_with_any_sample_ran":1,"distinct_papers_with_graph_line":2,"distinct_papers_with_any_sample_ran":1,"samples_over_distinct_papers":{"n_ran":1,"n_unverified":4,"n_samples":5,"n_pointer_only_licence":3,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":1,"n_unverified":4,"n_samples":5,"n_pointer_only_licence":3,"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"}}}