{"url":"/sota/skeleton-based-action-recognition-on-n-ucla","task":{"name":"Skeleton Based Action Recognition","url":"/task/skeleton-based-action-recognition","note":null},"dataset":{"name":"N-UCLA","url":"/dataset/n-ucla"},"category":"Computer Vision","categories":["Computer Vision","Natural Language Processing","Time Series"],"category_note":null,"description":"**Skeleton-based Action Recognition** is a computer vision task that involves recognizing human actions from a sequence of 3D skeletal joint data captured from sensors such as Microsoft Kinect, Intel RealSense, and wearable devices. The goal of skeleton-based action recognition is to develop algorithms that can understand and classify human actions from skeleton data, which can be used in various applications such as human-computer interaction, sports analysis, and surveillance.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [View Adaptive Neural Networks for High\r\nPerformance Skeleton-based Human Action\r\nRecognition](https://arxiv.org/pdf/1804.07453v3.pdf) )</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":["Accuracy","Data Modality (Joint, Bone, Motion)"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Accuracy":"higher","Data Modality (Joint, Bone, Motion)":null}},"counts":{"rows":25,"rows_with_code":21,"rows_with_paper_page":25,"rows_dated":25,"rows_using_additional_data":4},"rows":[{"rank_in_archive_order":1,"model":"DSCNet (RGB + Pose)","metrics":{"Accuracy":"99.1"},"uses_additional_data":false,"paper_date":"2023-12-28","paper":"/paper/a-dense-sparse-complementary-network-for","paper_url":"https://www.sciencedirect.com/science/article/pii/S0957417423035637","paper_title":"A Dense-Sparse Complementary Network for Human Action Recognition based on RGB and Skeleton Modalities","code":"https://github.com/Maxchengqin/DSCNet","n_code_links":1,"syntology":null},{"rank_in_archive_order":2,"model":"SkateFormer","metrics":{"Accuracy":"98.3"},"uses_additional_data":false,"paper_date":"2024-03-14","paper":"/paper/skateformer-skeletal-temporal-transformer-for","paper_url":"https://arxiv.org/abs/2403.09508v3","paper_title":"SkateFormer: Skeletal-Temporal Transformer for Human Action Recognition","code":"https://github.com/KAIST-VICLab/SkateFormer","n_code_links":1,"syntology":{"n_ran":3,"n_unverified":0,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":3,"model":"LA-GCN","metrics":{"Accuracy":"97.6"},"uses_additional_data":false,"paper_date":"2023-05-21","paper":"/paper/language-knowledge-assisted-representation","paper_url":"https://arxiv.org/abs/2305.12398v1","paper_title":"Language Knowledge-Assisted Representation Learning for Skeleton-Based Action Recognition","code":"https://github.com/damnull/lagcn","n_code_links":1,"syntology":null},{"rank_in_archive_order":4,"model":"MMCL","metrics":{"Accuracy":"97.5"},"uses_additional_data":true,"paper_date":"2024-07-22","paper":"/paper/multi-modality-co-learning-for-efficient-1","paper_url":"https://arxiv.org/abs/2407.15706v6","paper_title":"Multi-Modality Co-Learning for Efficient Skeleton-based Action Recognition","code":"https://github.com/liujf69/MMCL-Action","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":1,"n_samples":10,"n_pointer_only_licence":10}},{"rank_in_archive_order":5,"model":"TD-GCN","metrics":{"Accuracy":"97.4"},"uses_additional_data":false,"paper_date":"2023-05-01","paper":"/paper/temporal-decoupling-graph-convolutional","paper_url":"https://ieeexplore.ieee.org/abstract/document/10113233/","paper_title":"Temporal Decoupling Graph Convolutional Network for Skeleton-based Gesture Recognition","code":"https://github.com/liujf69/TD-GCN-Gesture","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"Action Capsules","metrics":{"Accuracy":"97.3"},"uses_additional_data":false,"paper_date":"2023-01-30","paper":"/paper/action-capsules-human-skeleton-action","paper_url":"https://arxiv.org/abs/2301.13090v1","paper_title":"Action Capsules: Human Skeleton Action Recognition","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":7,"model":"LST","metrics":{"Accuracy":"97.2"},"uses_additional_data":false,"paper_date":"2022-08-10","paper":"/paper/language-supervised-training-for-skeleton","paper_url":"https://arxiv.org/abs/2208.05318v2","paper_title":"Generative Action Description Prompts for Skeleton-based Action Recognition","code":"https://github.com/martinxm/gap","n_code_links":3,"syntology":{"n_ran":7,"n_unverified":3,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":8,"model":"HD-GCN","metrics":{"Accuracy":"97.2"},"uses_additional_data":false,"paper_date":"2022-08-23","paper":"/paper/hierarchically-decomposed-graph-convolutional","paper_url":"https://arxiv.org/abs/2208.10741v3","paper_title":"Hierarchically Decomposed Graph Convolutional Networks for Skeleton-Based Action Recognition","code":"https://github.com/Jho-Yonsei/HD-GCN","n_code_links":1,"syntology":{"n_ran":8,"n_unverified":3,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":9,"model":"InfoGCN","metrics":{"Accuracy":"97.0"},"uses_additional_data":false,"paper_date":"2022-01-01","paper":"/paper/infogcn-representation-learning-for-human","paper_url":"http://openaccess.thecvf.com//content/CVPR2022/html/Chi_InfoGCN_Representation_Learning_for_Human_Skeleton-Based_Action_Recognition_CVPR_2022_paper.html","paper_title":"InfoGCN: Representation Learning for Human Skeleton-Based Action Recognition","code":"https://github.com/stnoah1/infogcn","n_code_links":1,"syntology":null},{"rank_in_archive_order":10,"model":"TCA-GCN","metrics":{"Accuracy":"97.0"},"uses_additional_data":false,"paper_date":"2022-05-31","paper":"/paper/skeleton-based-action-recognition-via","paper_url":"https://arxiv.org/abs/2205.15936v2","paper_title":"Skeleton-based Action Recognition via Temporal-Channel Aggregation","code":"https://github.com/OrdinaryQin/TCA-GCN","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"DSTSA-GCN","metrics":{"Accuracy":"96.98"},"uses_additional_data":false,"paper_date":"2025-01-21","paper":"/paper/dstsa-gcn-advancing-skeleton-based-gesture","paper_url":"https://arxiv.org/abs/2501.12086v1","paper_title":"DSTSA-GCN: Advancing Skeleton-Based Gesture Recognition with Semantic-Aware Spatio-Temporal Topology Modeling","code":"https://github.com/HuCui2022/DSTSA-GCN_Gesture","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"JPFormer","metrics":{"Accuracy":"96.7"},"uses_additional_data":false,"paper_date":"2024-07-30","paper":"/paper/joint-partition-group-attention-for-skeleton","paper_url":"https://www.sciencedirect.com/science/article/pii/S0165168424002111","paper_title":"Joint-Partition Group Attention for skeleton-based action recognition","code":"https://github.com/HuCui2022/JPFormer","n_code_links":1,"syntology":null},{"rank_in_archive_order":13,"model":"CTR-GCN","metrics":{"Accuracy":"96.5"},"uses_additional_data":false,"paper_date":"2021-07-26","paper":"/paper/channel-wise-topology-refinement-graph","paper_url":"https://arxiv.org/abs/2107.12213v2","paper_title":"Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition","code":"https://github.com/kennymckormick/pyskl","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":3,"n_samples":5,"n_pointer_only_licence":3}},{"rank_in_archive_order":14,"model":"RE-TCN","metrics":{"Accuracy":"96.34%","Data Modality (Joint, Bone, Motion)":"Joint"},"uses_additional_data":false,"paper_date":"2025-03-29","paper":"/paper/action-recognition-in-real-world-ambient","paper_url":"https://arxiv.org/abs/2503.23214v1","paper_title":"Action Recognition in Real-World Ambient Assisted Living Environment","code":"https://github.com/Gbouna/RE-TCN","n_code_links":1,"syntology":null},{"rank_in_archive_order":15,"model":"TD-GDSCN","metrics":{"Accuracy":"95.69","Data Modality (Joint, Bone, Motion)":"Joint"},"uses_additional_data":false,"paper_date":"2024-08-15","paper":"/paper/action-recognition-for-privacy-preserving","paper_url":"https://www.researchgate.net/publication/383206363_Action_Recognition_for_Privacy-Preserving_Ambient_Assisted_Living","paper_title":"Action Recognition for Privacy-Preserving Ambient Assisted Living","code":"https://github.com/Gbouna/Action-Recognition-for-Privacy-Preserving-Ambient-Assisted-Living","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"MSSTNet","metrics":{"Accuracy":"95.3"},"uses_additional_data":false,"paper_date":"2023-05-12","paper":"/paper/multi-scale-spatial-temporal-convolutional","paper_url":"https://link.springer.com/article/10.1007/s10044-023-01156-w#Sec8","paper_title":"Multi-scale spatial–temporal convolutional neural network for skeleton-based action recognition","code":"https://github.com/Maxchengqin/MSSTNet_2D_3D","n_code_links":1,"syntology":null},{"rank_in_archive_order":17,"model":"Hierarchical Action Classification (RGB + Pose)","metrics":{"Accuracy":"93.99"},"uses_additional_data":false,"paper_date":"2020-07-30","paper":"/paper/hierarchical-action-classification-with","paper_url":"https://arxiv.org/abs/2007.15244v1","paper_title":"Hierarchical Action Classification with Network Pruning","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":18,"model":"MMNet (RGB + Pose)","metrics":{"Accuracy":"93.7"},"uses_additional_data":true,"paper_date":"2022-05-26","paper":"/paper/mmnet-a-model-based-multimodal-network-for","paper_url":"https://ieeexplore.ieee.org/abstract/document/9782511","paper_title":"MMNet: A Model-Based Multimodal Network for Human Action Recognition in RGB-D Videos","code":"https://github.com/bruceyo/MMNet","n_code_links":1,"syntology":null},{"rank_in_archive_order":19,"model":"VPN (RGB + Pose)","metrics":{"Accuracy":"93.5"},"uses_additional_data":true,"paper_date":"2020-07-06","paper":"/paper/vpn-learning-video-pose-embedding-for","paper_url":"https://arxiv.org/abs/2007.03056v1","paper_title":"VPN: Learning Video-Pose Embedding for Activities of Daily Living","code":"https://github.com/srijandas07/VPN","n_code_links":1,"syntology":null},{"rank_in_archive_order":20,"model":"VPN++ (RGB + Pose)","metrics":{"Accuracy":"93.5"},"uses_additional_data":true,"paper_date":"2021-05-17","paper":"/paper/vpn-rethinking-video-pose-embeddings-for","paper_url":"https://arxiv.org/abs/2105.08141v1","paper_title":"VPN++: Rethinking Video-Pose embeddings for understanding Activities of Daily Living","code":"https://github.com/srijandas07/vpnplusplus","n_code_links":1,"syntology":{"n_ran":1,"n_unverified":0,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":21,"model":"SGN","metrics":{"Accuracy":"92.5%"},"uses_additional_data":false,"paper_date":"2019-04-02","paper":"/paper/semantics-guided-neural-networks-for","paper_url":"https://arxiv.org/abs/1904.01189v3","paper_title":"Semantics-Guided Neural Networks for Efficient Skeleton-Based Human Action Recognition","code":"https://github.com/microsoft/SGN","n_code_links":2,"syntology":{"n_ran":3,"n_unverified":10,"n_samples":13,"n_pointer_only_licence":0}},{"rank_in_archive_order":22,"model":"Action Machine","metrics":{"Accuracy":"92.3%"},"uses_additional_data":false,"paper_date":"2018-12-14","paper":"/paper/action-machine-rethinking-action-recognition","paper_url":"http://arxiv.org/abs/1812.05770v2","paper_title":"Action Machine: Rethinking Action Recognition in Trimmed Videos","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":23,"model":"EleAtt-GRU (aug.)","metrics":{"Accuracy":"90.7%"},"uses_additional_data":false,"paper_date":"2019-09-03","paper":"/paper/eleatt-rnn-adding-attentiveness-to-neurons-in","paper_url":"https://arxiv.org/abs/1909.01939v1","paper_title":"EleAtt-RNN: Adding Attentiveness to Neurons in Recurrent Neural Networks","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":24,"model":"VA-fusion (aug.)","metrics":{"Accuracy":"88.1%"},"uses_additional_data":false,"paper_date":"2018-04-20","paper":"/paper/view-adaptive-neural-networks-for-high","paper_url":"https://arxiv.org/abs/1804.07453v3","paper_title":"View Adaptive Neural Networks for High Performance Skeleton-based Human Action Recognition","code":"https://github.com/microsoft/View-Adaptive-Neural-Networks-for-Skeleton-based-Human-Action-Recognition","n_code_links":2,"syntology":{"n_ran":6,"n_unverified":6,"n_samples":12,"n_pointer_only_licence":9}},{"rank_in_archive_order":25,"model":"Glimpse Clouds","metrics":{"Accuracy":"87.6%"},"uses_additional_data":false,"paper_date":"2018-02-22","paper":"/paper/glimpse-clouds-human-activity-recognition","paper_url":"http://arxiv.org/abs/1802.07898v4","paper_title":"Glimpse Clouds: Human Activity Recognition from Unstructured Feature Points","code":"https://github.com/fabienbaradel/glimpse_clouds","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":8,"rows_with_any_sample_ran":8,"distinct_papers_with_graph_line":8,"distinct_papers_with_any_sample_ran":8,"samples_over_distinct_papers":{"n_ran":39,"n_unverified":26,"n_samples":65,"n_pointer_only_licence":23,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":39,"n_unverified":26,"n_samples":65,"n_pointer_only_licence":23,"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"}}}