{"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/cascaded-boundary-regression-for-temporal","title":"Cascaded Boundary Regression for Temporal Action Detection","arxiv_id":"1705.01180","date":"2017-05-02","proceeding":null,"authors":["Jiyang Gao","Zhenheng Yang","Ram Nevatia"],"abstract":"Temporal action detection in long videos is an important problem.\nState-of-the-art methods address this problem by applying action classifiers on\nsliding windows. Although sliding windows may contain an identifiable portion\nof the actions, they may not necessarily cover the entire action instance,\nwhich would lead to inferior performance. We adapt a two-stage temporal action\ndetection pipeline with Cascaded Boundary Regression (CBR) model.\nClass-agnostic proposals and specific actions are detected respectively in the\nfirst and the second stage. CBR uses temporal coordinate regression to refine\nthe temporal boundaries of the sliding windows. The salient aspect of the\nrefinement process is that, inside each stage, the temporal boundaries are\nadjusted in a cascaded way by feeding the refined windows back to the system\nfor further boundary refinement. We test CBR on THUMOS-14 and TVSeries, and\nachieve state-of-the-art performance on both datasets. The performance gain is\nespecially remarkable under high IoU thresholds, e.g. map@tIoU=0.5 on THUMOS-14\nis improved from 19.0% to 31.0%.","url_abs":"http://arxiv.org/abs/1705.01180v1","url_pdf":"http://arxiv.org/pdf/1705.01180v1.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-detection","task_name":"Action Detection"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/temporal-action-localization-on-thumos14","task":"Temporal Action Localization","dataset":"THUMOS’14","model":"CBR-TS","rank_in_archive_order":34,"of":42,"metrics":{"mAP IOU@0.1":"60.1","mAP IOU@0.2":"56.7","mAP IOU@0.3":"50.1","mAP IOU@0.4":"41.3","mAP IOU@0.5":"31","mAP IOU@0.6":"19.1","mAP IOU@0.7":"9.9"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1705.01180","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}