{"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-segment-networks-for-action","title":"Temporal Segment Networks for Action Recognition in Videos","arxiv_id":"1705.02953","date":"2017-05-08","proceeding":null,"authors":["Limin Wang","Yuanjun Xiong","Zhe Wang","Yu Qiao","Dahua Lin","Xiaoou Tang","Luc van Gool"],"abstract":"Deep convolutional networks have achieved great success for image\nrecognition. However, for action recognition in videos, their advantage over\ntraditional methods is not so evident. We present a general and flexible\nvideo-level framework for learning action models in videos. This method, called\ntemporal segment network (TSN), aims to model long-range temporal structures\nwith a new segment-based sampling and aggregation module. This unique design\nenables our TSN to efficiently learn action models by using the whole action\nvideos. The learned models could be easily adapted for action recognition in\nboth trimmed and untrimmed videos with simple average pooling and multi-scale\ntemporal window integration, respectively. We also study a series of good\npractices for the instantiation of TSN framework given limited training\nsamples. Our approach obtains the state-the-of-art performance on four\nchallenging action recognition benchmarks: HMDB51 (71.0%), UCF101 (94.9%),\nTHUMOS14 (80.1%), and ActivityNet v1.2 (89.6%). Using the proposed RGB\ndifference for motion models, our method can still achieve competitive accuracy\non UCF101 (91.0%) while running at 340 FPS. Furthermore, based on the temporal\nsegment networks, we won the video classification track at the ActivityNet\nchallenge 2016 among 24 teams, which demonstrates the effectiveness of TSN and\nthe proposed good practices.","url_abs":"http://arxiv.org/abs/1705.02953v1","url_pdf":"http://arxiv.org/pdf/1705.02953v1.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-segment-networks-for-action","repo_url":"https://github.com/yjxiong/temporal-segment-networks","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"temporal-segment-networks-for-action","repo_url":"https://github.com/yjxiong/caffe","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"temporal-segment-networks-for-action","repo_url":"https://github.com/ayushrox/TSN_Colab","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"temporal-segment-networks-for-action","repo_url":"https://github.com/mtlouie-unm/alome-tsn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"temporal-segment-networks-for-action","repo_url":"https://github.com/peachman05/action-recognition-tutorial","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"temporal-segment-networks-for-action","repo_url":"https://github.com/thaitruongan/camera-surveillance-ai","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"temporal-segment-networks-for-action","repo_url":"https://github.com/2023-MindSpore-1/ms-code-7/tree/main/tsn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"temporal-segment-networks-for-action","repo_url":"https://github.com/MindSpore-paper-code-2/code3/tree/main/tsn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"temporal-segment-networks-for-action","repo_url":"https://github.com/PaddlePaddle/PaddleVideo/blob/develop/docs/zh-CN/model_zoo/recognition/pp-tsn.md","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":null},{"paper_slug":"temporal-segment-networks-for-action","repo_url":"https://github.com/open-mmlab/mmaction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"temporal-segment-networks-for-action","repo_url":"https://github.com/open-mmlab/mmaction2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"action-classification","task_name":"Action Classification"},{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"action-recognition-in-videos-2","task_name":"Action Recognition In Videos"},{"task_slug":"micro-action-recognition","task_name":"Micro-Action Recognition"},{"task_slug":"action-recognition","task_name":"Temporal Action Localization"},{"task_slug":"video-classification","task_name":"Video Classification"}],"methods":[{"method_slug":"average-pooling","method_name":"Average Pooling"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/action-classification-on-moments-in-time","task":"Action Classification","dataset":"MiT","model":"TSN-2Stream","rank_in_archive_order":29,"of":29,"metrics":{"Top 5 Accuracy":"50.10%"},"uses_additional_data":false},{"leaderboard":"/sota/video-classification-on-coin-1","task":"Video Classification","dataset":"COIN","model":"TSN","rank_in_archive_order":7,"of":7,"metrics":{"Accuracy (%)":"73.4"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1705.02953","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}