{"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/temporal-action-localization-with-enhanced","title":"Temporal Action Localization with Enhanced Instant Discriminability","arxiv_id":"2309.05590","date":"2023-09-11","proceeding":null,"authors":["Dingfeng Shi","Qiong Cao","Yujie Zhong","Shan An","Jian Cheng","Haogang Zhu","DaCheng Tao"],"abstract":"Temporal action detection (TAD) aims to detect all action boundaries and their corresponding categories in an untrimmed video. The unclear boundaries of actions in videos often result in imprecise predictions of action boundaries by existing methods. To resolve this issue, we propose a one-stage framework named TriDet. First, we propose a Trident-head to model the action boundary via an estimated relative probability distribution around the boundary. Then, we analyze the rank-loss problem (i.e. instant discriminability deterioration) in transformer-based methods and propose an efficient scalable-granularity perception (SGP) layer to mitigate this issue. To further push the limit of instant discriminability in the video backbone, we leverage the strong representation capability of pretrained large models and investigate their performance on TAD. Last, considering the adequate spatial-temporal context for classification, we design a decoupled feature pyramid network with separate feature pyramids to incorporate rich spatial context from the large model for localization. Experimental results demonstrate the robustness of TriDet and its state-of-the-art performance on multiple TAD datasets, including hierarchical (multilabel) TAD datasets.","url_abs":"https://arxiv.org/abs/2309.05590v1","url_pdf":"https://arxiv.org/pdf/2309.05590v1.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":"temporal-action-localization-with-enhanced","repo_url":"https://github.com/dingfengshi/tridetplus","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"temporal-action-localization-with-enhanced","repo_url":"https://github.com/dingfengshi/tridet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"temporal-action-localization-with-enhanced","repo_url":"https://github.com/sssste/tridet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"action-detection","task_name":"Action Detection"},{"task_slug":"action-localization","task_name":"Action Localization"},{"task_slug":"action-recognition","task_name":"Temporal Action Localization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/temporal-action-localization-on-hacs","task":"Temporal Action Localization","dataset":"HACS","model":"TriDet (VideoMAEv2)","rank_in_archive_order":5,"of":12,"metrics":{"Average-mAP":"43.1","mAP@0.5":"62.4","mAP@0.75":"44.1","mAP@0.95":"13.1"},"uses_additional_data":false},{"leaderboard":"/sota/temporal-action-localization-on-multithumos-1","task":"Temporal Action Localization","dataset":"MultiTHUMOS","model":"TriDet (VideoMAEv2)","rank_in_archive_order":1,"of":8,"metrics":{"Average mAP":"37.5","mAP IOU@0.2":"57.7","mAP IOU@0.5":"42.7","mAP IOU@0.7":"24.3"},"uses_additional_data":false},{"leaderboard":"/sota/temporal-action-localization-on-multithumos-1","task":"Temporal Action Localization","dataset":"MultiTHUMOS","model":"TriDet (I3D-rgb)","rank_in_archive_order":3,"of":8,"metrics":{"Average mAP":"30.7","mAP IOU@0.2":"49.1","mAP IOU@0.5":"34.3","mAP IOU@0.7":"17.8"},"uses_additional_data":false},{"leaderboard":"/sota/temporal-action-localization-on-thumos14","task":"Temporal Action Localization","dataset":"THUMOS’14","model":"TriDet (VideoMAE v2-g feature)","rank_in_archive_order":6,"of":42,"metrics":{"Avg mAP (0.3:0.7)":"70.1","mAP IOU@0.3":"84.8","mAP IOU@0.4":"80.0","mAP IOU@0.5":"73.3","mAP IOU@0.6":"63.8","mAP IOU@0.7":"48.8"},"uses_additional_data":true}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2309.05590","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}