{"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/memory-augmented-attention-modelling-for","title":"Memory-augmented Attention Modelling for Videos","arxiv_id":"1611.02261","date":"2016-11-07","proceeding":null,"authors":["Rasool Fakoor","Abdel-rahman Mohamed","Margaret Mitchell","Sing Bing Kang","Pushmeet Kohli"],"abstract":"We present a method to improve video description generation by modeling\nhigher-order interactions between video frames and described concepts. By\nstoring past visual attention in the video associated to previously generated\nwords, the system is able to decide what to look at and describe in light of\nwhat it has already looked at and described. This enables not only more\neffective local attention, but tractable consideration of the video sequence\nwhile generating each word. Evaluation on the challenging and popular MSVD and\nCharades datasets demonstrates that the proposed architecture outperforms\nprevious video description approaches without requiring external temporal video\nfeatures.","url_abs":"http://arxiv.org/abs/1611.02261v4","url_pdf":"http://arxiv.org/pdf/1611.02261v4.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":"memory-augmented-attention-modelling-for","repo_url":"https://github.com/rasoolfa/videocap","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"torch","reach":null}],"tasks":[{"task_slug":"video-description","task_name":"Video Description"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}