{"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/from-trailers-to-storylines-an-efficient-way","title":"From Trailers to Storylines: An Efficient Way to Learn from Movies","arxiv_id":"1806.05341","date":"2018-06-14","proceeding":null,"authors":["Qingqiu Huang","Yuanjun Xiong","Yu Xiong","Yuqi Zhang","Dahua Lin"],"abstract":"The millions of movies produced in the human history are valuable resources\nfor computer vision research. However, learning a vision model from movie data\nwould meet with serious difficulties. A major obstacle is the computational\ncost -- the length of a movie is often over one hour, which is substantially\nlonger than the short video clips that previous study mostly focuses on. In\nthis paper, we explore an alternative approach to learning vision models from\nmovies. Specifically, we consider a framework comprised of a visual module and\na temporal analysis module. Unlike conventional learning methods, the proposed\napproach learns these modules from different sets of data -- the former from\ntrailers while the latter from movies. This allows distinctive visual features\nto be learned within a reasonable budget while still preserving long-term\ntemporal structures across an entire movie. We construct a large-scale dataset\nfor this study and define a series of tasks on top. Experiments on this dataset\nshowed that the proposed method can substantially reduce the training time\nwhile obtaining highly effective features and coherent temporal structures.","url_abs":"http://arxiv.org/abs/1806.05341v1","url_pdf":"http://arxiv.org/pdf/1806.05341v1.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":"from-trailers-to-storylines-an-efficient-way","repo_url":"https://github.com/ycxioooong/MovieSynopsisAssociation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.05341","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}