{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/dg-stgcn-dynamic-spatial-temporal-modeling","title":"DG-STGCN: Dynamic Spatial-Temporal Modeling for Skeleton-based Action Recognition","arxiv_id":"2210.05895","date":"2022-10-12","proceeding":null,"authors":["Haodong Duan","Jiaqi Wang","Kai Chen","Dahua Lin"],"abstract":"Graph convolution networks (GCN) have been widely used in skeleton-based action recognition. We note that existing GCN-based approaches primarily rely on prescribed graphical structures (ie., a manually defined topology of skeleton joints), which limits their flexibility to capture complicated correlations between joints. To move beyond this limitation, we propose a new framework for skeleton-based action recognition, namely Dynamic Group Spatio-Temporal GCN (DG-STGCN). It consists of two modules, DG-GCN and DG-TCN, respectively, for spatial and temporal modeling. In particular, DG-GCN uses learned affinity matrices to capture dynamic graphical structures instead of relying on a prescribed one, while DG-TCN performs group-wise temporal convolutions with varying receptive fields and incorporates a dynamic joint-skeleton fusion module for adaptive multi-level temporal modeling. On a wide range of benchmarks, including NTURGB+D, Kinetics-Skeleton, BABEL, and Toyota SmartHome, DG-STGCN consistently outperforms state-of-the-art methods, often by a notable margin.","url_abs":"https://arxiv.org/abs/2210.05895v1","url_pdf":"https://arxiv.org/pdf/2210.05895v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"dg-stgcn-dynamic-spatial-temporal-modeling","repo_url":"https://github.com/kennymckormick/pyskl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"dg-stgcn-dynamic-spatial-temporal-modeling","repo_url":"https://github.com/2023-MindSpore-1/ms-code-217/tree/main/stgcn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"dg-stgcn-dynamic-spatial-temporal-modeling","repo_url":"https://github.com/MindSpore-paper-code-2/code3/tree/main/stgcn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"skeleton-based-action-recognition","task_name":"Skeleton Based Action Recognition"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"gcn","method_name":"GCN"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/skeleton-based-action-recognition-on-ntu-rgbd","task":"Skeleton Based Action Recognition","dataset":"NTU RGB+D","model":"DG-STGCN","rank_in_archive_order":15,"of":135,"metrics":{"Accuracy (CS)":"93.2","Accuracy (CV)":"97.5","Ensembled Modalities":"4"},"uses_additional_data":false},{"leaderboard":"/sota/skeleton-based-action-recognition-on-ntu-rgbd-1","task":"Skeleton Based Action Recognition","dataset":"NTU RGB+D 120","model":"DG-STGCN","rank_in_archive_order":17,"of":83,"metrics":{"Accuracy (Cross-Setup)":"91.3","Accuracy (Cross-Subject)":"89.6","Ensembled Modalities":"4"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2210.05895","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}