{"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/encouraging-lstms-to-anticipate-actions-very","title":"Encouraging LSTMs to Anticipate Actions Very Early","arxiv_id":"1703.07023","date":"2017-03-21","proceeding":"ICCV 2017 10","authors":["Mohammad Sadegh Aliakbarian","Fatemeh Sadat Saleh","Mathieu Salzmann","Basura Fernando","Lars Petersson","Lars Andersson"],"abstract":"In contrast to the widely studied problem of recognizing an action given a\ncomplete sequence, action anticipation aims to identify the action from only\npartially available videos. As such, it is therefore key to the success of\ncomputer vision applications requiring to react as early as possible, such as\nautonomous navigation. In this paper, we propose a new action anticipation\nmethod that achieves high prediction accuracy even in the presence of a very\nsmall percentage of a video sequence. To this end, we develop a multi-stage\nLSTM architecture that leverages context-aware and action-aware features, and\nintroduce a novel loss function that encourages the model to predict the\ncorrect class as early as possible. Our experiments on standard benchmark\ndatasets evidence the benefits of our approach; We outperform the\nstate-of-the-art action anticipation methods for early prediction by a relative\nincrease in accuracy of 22.0% on JHMDB-21, 14.0% on UT-Interaction and 49.9% on\nUCF-101.","url_abs":"http://arxiv.org/abs/1703.07023v3","url_pdf":"http://arxiv.org/pdf/1703.07023v3.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":"encouraging-lstms-to-anticipate-actions-very","repo_url":"https://github.com/mangalutsav/Multi-Stage-LSTM-for-Action-Anticipation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"action-anticipation","task_name":"Action Anticipation"},{"task_slug":"autonomous-navigation","task_name":"Autonomous Navigation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1703.07023","atlas_url":"https://app.syntology.ai/?focus=1703.07023","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}