{"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-regularity-in-skeleton-trajectories","title":"Learning Regularity in Skeleton Trajectories for Anomaly Detection in Videos","arxiv_id":"1903.03295","date":"2019-03-08","proceeding":"CVPR 2019 6","authors":["Romero Morais","Vuong Le","Truyen Tran","Budhaditya Saha","Moussa Mansour","Svetha Venkatesh"],"abstract":"Appearance features have been widely used in video anomaly detection even\nthough they contain complex entangled factors. We propose a new method to model\nthe normal patterns of human movements in surveillance video for anomaly\ndetection using dynamic skeleton features. We decompose the skeletal movements\ninto two sub-components: global body movement and local body posture. We model\nthe dynamics and interaction of the coupled features in our novel\nMessage-Passing Encoder-Decoder Recurrent Network. We observed that the\ndecoupled features collaboratively interact in our spatio-temporal model to\naccurately identify human-related irregular events from surveillance video\nsequences. Compared to traditional appearance-based models, our method achieves\nsuperior outlier detection performance. Our model also offers \"open-box\"\nexamination and decision explanation made possible by the semantically\nunderstandable features and a network architecture supporting interpretability.","url_abs":"http://arxiv.org/abs/1903.03295v2","url_pdf":"http://arxiv.org/pdf/1903.03295v2.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-regularity-in-skeleton-trajectories","repo_url":"https://github.com/RomeroBarata/skeleton_based_anomaly_detection","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"anomaly-detection","task_name":"Anomaly Detection"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"outlier-detection","task_name":"Outlier Detection"},{"task_slug":"video-anomaly-detection","task_name":"Video Anomaly Detection"}],"methods":[],"datasets_introduced":[{"slug":"hr-avenue","name":"HR-Avenue","full_name":""},{"slug":"hr-shanghaitech","name":"HR-ShanghaiTech","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/anomaly-detection-on-shanghaitech","task":"Anomaly Detection","dataset":"ShanghaiTech","model":"MPED-RNN","rank_in_archive_order":27,"of":31,"metrics":{"AUC":"73.40%"},"uses_additional_data":false},{"leaderboard":"/sota/video-anomaly-detection-on-hr-avenue","task":"Video Anomaly Detection","dataset":"HR-Avenue","model":"MPED-RNN","rank_in_archive_order":7,"of":11,"metrics":{"AUC":"86.3"},"uses_additional_data":false},{"leaderboard":"/sota/video-anomaly-detection-on-hr-shanghaitech","task":"Video Anomaly Detection","dataset":"HR-ShanghaiTech","model":"MPED-RNN","rank_in_archive_order":9,"of":14,"metrics":{"AUC":"75.4"},"uses_additional_data":false},{"leaderboard":"/sota/video-anomaly-detection-on-hr-ubnormal","task":"Video Anomaly Detection","dataset":"HR-UBnormal","model":"MPED-RNN","rank_in_archive_order":6,"of":8,"metrics":{"AUC":"61.2"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.03295","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}