{"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/structure-inference-machines-recurrent-neural","title":"Structure Inference Machines: Recurrent Neural Networks for Analyzing Relations in Group Activity Recognition","arxiv_id":"1511.04196","date":"2015-11-13","proceeding":"CVPR 2016 6","authors":["Zhiwei Deng","Arash Vahdat","Hexiang Hu","Greg Mori"],"abstract":"Rich semantic relations are important in a variety of visual recognition\nproblems. As a concrete example, group activity recognition involves the\ninteractions and relative spatial relations of a set of people in a scene.\nState of the art recognition methods center on deep learning approaches for\ntraining highly effective, complex classifiers for interpreting images.\nHowever, bridging the relatively low-level concepts output by these methods to\ninterpret higher-level compositional scenes remains a challenge. Graphical\nmodels are a standard tool for this task. In this paper, we propose a method to\nintegrate graphical models and deep neural networks into a joint framework.\nInstead of using a traditional inference method, we use a sequential inference\nmodeled by a recurrent neural network. Beyond this, the appropriate structure\nfor inference can be learned by imposing gates on edges between nodes.\nEmpirical results on group activity recognition demonstrate the potential of\nthis model to handle highly structured learning tasks.","url_abs":"http://arxiv.org/abs/1511.04196v2","url_pdf":"http://arxiv.org/pdf/1511.04196v2.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":"activity-recognition","task_name":"Activity Recognition"},{"task_slug":"group-activity-recognition","task_name":"Group Activity Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/group-activity-recognition-on-collective","task":"Group Activity Recognition","dataset":"Collective Activity","model":"Deng et al.","rank_in_archive_order":6,"of":6,"metrics":{"Accuracy":"81.2"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1511.04196","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}