{"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/hidden-two-stream-convolutional-networks-for","title":"Hidden Two-Stream Convolutional Networks for Action Recognition","arxiv_id":"1704.00389","date":"2017-04-02","proceeding":null,"authors":["Yi Zhu","Zhenzhong Lan","Shawn Newsam","Alexander G. Hauptmann"],"abstract":"Analyzing videos of human actions involves understanding the temporal\nrelationships among video frames. State-of-the-art action recognition\napproaches rely on traditional optical flow estimation methods to pre-compute\nmotion information for CNNs. Such a two-stage approach is computationally\nexpensive, storage demanding, and not end-to-end trainable. In this paper, we\npresent a novel CNN architecture that implicitly captures motion information\nbetween adjacent frames. We name our approach hidden two-stream CNNs because it\nonly takes raw video frames as input and directly predicts action classes\nwithout explicitly computing optical flow. Our end-to-end approach is 10x\nfaster than its two-stage baseline. Experimental results on four challenging\naction recognition datasets: UCF101, HMDB51, THUMOS14 and ActivityNet v1.2 show\nthat our approach significantly outperforms the previous best real-time\napproaches.","url_abs":"http://arxiv.org/abs/1704.00389v4","url_pdf":"http://arxiv.org/pdf/1704.00389v4.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":"hidden-two-stream-convolutional-networks-for","repo_url":"https://github.com/bryanyzhu/Hidden-Two-Stream","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"hidden-two-stream-convolutional-networks-for","repo_url":"https://github.com/AbdalaDiasse/Video-classification-for-oil-quality-estimation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"hidden-two-stream-convolutional-networks-for","repo_url":"https://github.com/bryanyzhu/two-stream-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"optical-flow-estimation","task_name":"Optical Flow Estimation"},{"task_slug":"action-recognition","task_name":"Temporal Action Localization"},{"task_slug":"two","task_name":"Vocal Bursts Valence Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/action-recognition-in-videos-on-hmdb-51","task":"Action Recognition","dataset":"HMDB-51","model":"Hidden Two-Stream","rank_in_archive_order":27,"of":77,"metrics":{"Average accuracy of 3 splits":"78.7"},"uses_additional_data":false},{"leaderboard":"/sota/action-recognition-in-videos-on-ucf101","task":"Action Recognition","dataset":"UCF101","model":"Hidden Two-Stream","rank_in_archive_order":23,"of":91,"metrics":{"3-fold Accuracy":"97.1"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.00389","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}