{"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/mac-mining-activity-concepts-for-language","title":"MAC: Mining Activity Concepts for Language-based Temporal Localization","arxiv_id":"1811.08925","date":"2018-11-21","proceeding":null,"authors":["Runzhou Ge","JIyang Gao","Kan Chen","Ram Nevatia"],"abstract":"We address the problem of language-based temporal localization in untrimmed\nvideos. Compared to temporal localization with fixed categories, this problem\nis more challenging as the language-based queries not only have no pre-defined\nactivity list but also may contain complex descriptions. Previous methods\naddress the problem by considering features from video sliding windows and\nlanguage queries and learning a subspace to encode their correlation, which\nignore rich semantic cues about activities in videos and queries. We propose to\nmine activity concepts from both video and language modalities by applying the\nactionness score enhanced Activity Concepts based Localizer (ACL).\nSpecifically, the novel ACL encodes the semantic concepts from verb-obj pairs\nin language queries and leverages activity classifiers' prediction scores to\nencode visual concepts. Besides, ACL also has the capability to regress sliding\nwindows as localization results. Experiments show that ACL significantly\noutperforms state-of-the-arts under the widely used metric, with more than 5%\nincrease on both Charades-STA and TACoS datasets.","url_abs":"http://arxiv.org/abs/1811.08925v1","url_pdf":"http://arxiv.org/pdf/1811.08925v1.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":"mac-mining-activity-concepts-for-language","repo_url":"https://github.com/runzhouge/MAC","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"mac-mining-activity-concepts-for-language","repo_url":"https://github.com/WuJie1010/Temporally-language-grounding","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"mac-mining-activity-concepts-for-language","repo_url":"https://github.com/madhawav/MML","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"language-based-temporal-localization","task_name":"Language-Based Temporal Localization"},{"task_slug":"temporal-localization","task_name":"Temporal Localization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.08925","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}