{"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/learning-actor-relation-graphs-for-group","title":"Learning Actor Relation Graphs for Group Activity Recognition","arxiv_id":"1904.10117","date":"2019-04-23","proceeding":"CVPR 2019 6","authors":["Jianchao Wu","Li-Min Wang","Li Wang","Jie Guo","Gangshan Wu"],"abstract":"Modeling relation between actors is important for recognizing group activity\nin a multi-person scene. This paper aims at learning discriminative relation\nbetween actors efficiently using deep models. To this end, we propose to build\na flexible and efficient Actor Relation Graph (ARG) to simultaneously capture\nthe appearance and position relation between actors. Thanks to the Graph\nConvolutional Network, the connections in ARG could be automatically learned\nfrom group activity videos in an end-to-end manner, and the inference on ARG\ncould be efficiently performed with standard matrix operations. Furthermore, in\npractice, we come up with two variants to sparsify ARG for more effective\nmodeling in videos: spatially localized ARG and temporal randomized ARG. We\nperform extensive experiments on two standard group activity recognition\ndatasets: the Volleyball dataset and the Collective Activity dataset, where\nstate-of-the-art performance is achieved on both datasets. We also visualize\nthe learned actor graphs and relation features, which demonstrate that the\nproposed ARG is able to capture the discriminative relation information for\ngroup activity recognition.","url_abs":"http://arxiv.org/abs/1904.10117v1","url_pdf":"http://arxiv.org/pdf/1904.10117v1.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":"learning-actor-relation-graphs-for-group","repo_url":"https://github.com/wjchaoGit/Group-Activity-Recognition","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"learning-actor-relation-graphs-for-group","repo_url":"https://github.com/East-Tree/groupAc_GCN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"activity-recognition","task_name":"Activity Recognition"},{"task_slug":"group-activity-recognition","task_name":"Group Activity Recognition"},{"task_slug":null,"task_name":"Relation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/group-activity-recognition-on-collective","task":"Group Activity Recognition","dataset":"Collective Activity","model":"GT (Inception-v3)","rank_in_archive_order":3,"of":6,"metrics":{"Accuracy":"91"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.10117","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}