{"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/query-adaptive-video-summarization-via","title":"Query-adaptive Video Summarization via Quality-aware Relevance Estimation","arxiv_id":"1705.00581","date":"2017-05-01","proceeding":null,"authors":["Arun Balajee Vasudevan","Michael Gygli","Anna Volokitin","Luc van Gool"],"abstract":"Although the problem of automatic video summarization has recently received a\nlot of attention, the problem of creating a video summary that also highlights\nelements relevant to a search query has been less studied. We address this\nproblem by posing query-relevant summarization as a video frame subset\nselection problem, which lets us optimise for summaries which are\nsimultaneously diverse, representative of the entire video, and relevant to a\ntext query. We quantify relevance by measuring the distance between frames and\nqueries in a common textual-visual semantic embedding space induced by a neural\nnetwork. In addition, we extend the model to capture query-independent\nproperties, such as frame quality. We compare our method against previous state\nof the art on textual-visual embeddings for thumbnail selection and show that\nour model outperforms them on relevance prediction. Furthermore, we introduce a\nnew dataset, annotated with diversity and query-specific relevance labels. On\nthis dataset, we train and test our complete model for video summarization and\nshow that it outperforms standard baselines such as Maximal Marginal Relevance.","url_abs":"http://arxiv.org/abs/1705.00581v2","url_pdf":"http://arxiv.org/pdf/1705.00581v2.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":"query-adaptive-video-summarization-via","repo_url":"https://github.com/arunbalajeev/query-video-summary","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"video-summarization","task_name":"Video Summarization"}],"methods":[],"datasets_introduced":[{"slug":"rad","name":"RAD","full_name":"RELEVANCE AND DIVERSITY DATASET"}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1705.00581","atlas_url":"https://app.syntology.ai/?focus=1705.00581","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}