{"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/spatio-temporal-graph-convolution-for","title":"Spatio-Temporal Graph Convolution for Skeleton Based Action Recognition","arxiv_id":"1802.09834","date":"2018-02-27","proceeding":null,"authors":["Chaolong Li","Zhen Cui","Wenming Zheng","Chunyan Xu","Jian Yang"],"abstract":"Variations of human body skeletons may be considered as dynamic graphs, which\nare generic data representation for numerous real-world applications. In this\npaper, we propose a spatio-temporal graph convolution (STGC) approach for\nassembling the successes of local convolutional filtering and sequence learning\nability of autoregressive moving average. To encode dynamic graphs, the\nconstructed multi-scale local graph convolution filters, consisting of matrices\nof local receptive fields and signal mappings, are recursively performed on\nstructured graph data of temporal and spatial domain. The proposed model is\ngeneric and principled as it can be generalized into other dynamic models. We\ntheoretically prove the stability of STGC and provide an upper-bound of the\nsignal transformation to be learnt. Further, the proposed recursive model can\nbe stacked into a multi-layer architecture. To evaluate our model, we conduct\nextensive experiments on four benchmark skeleton-based action datasets,\nincluding the large-scale challenging NTU RGB+D. The experimental results\ndemonstrate the effectiveness of our proposed model and the improvement over\nthe state-of-the-art.","url_abs":"http://arxiv.org/abs/1802.09834v1","url_pdf":"http://arxiv.org/pdf/1802.09834v1.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":[],"tasks":[{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"skeleton-based-action-recognition","task_name":"Skeleton Based Action Recognition"},{"task_slug":"action-recognition","task_name":"Temporal Action Localization"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/skeleton-based-action-recognition-on-florence","task":"Skeleton Based Action Recognition","dataset":"Florence 3D","model":"Deep STGC_K","rank_in_archive_order":1,"of":7,"metrics":{"Accuracy":"99.1%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.09834","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}