{"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/objects2action-classifying-and-localizing","title":"Objects2action: Classifying and localizing actions without any video example","arxiv_id":"1510.06939","date":"2015-10-23","proceeding":"ICCV 2015 12","authors":["Mihir Jain","Jan C. van Gemert","Thomas Mensink","Cees G. M. Snoek"],"abstract":"The goal of this paper is to recognize actions in video without the need for\nexamples. Different from traditional zero-shot approaches we do not demand the\ndesign and specification of attribute classifiers and class-to-attribute\nmappings to allow for transfer from seen classes to unseen classes. Our key\ncontribution is objects2action, a semantic word embedding that is spanned by a\nskip-gram model of thousands of object categories. Action labels are assigned\nto an object encoding of unseen video based on a convex combination of action\nand object affinities. Our semantic embedding has three main characteristics to\naccommodate for the specifics of actions. First, we propose a mechanism to\nexploit multiple-word descriptions of actions and objects. Second, we\nincorporate the automated selection of the most responsive objects per action.\nAnd finally, we demonstrate how to extend our zero-shot approach to the\nspatio-temporal localization of actions in video. Experiments on four action\ndatasets demonstrate the potential of our approach.","url_abs":"http://arxiv.org/abs/1510.06939v1","url_pdf":"http://arxiv.org/pdf/1510.06939v1.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":[],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"object","task_name":"Object"},{"task_slug":"temporal-localization","task_name":"Temporal Localization"},{"task_slug":"zero-shot-action-recognition","task_name":"Zero-Shot Action Recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/zero-shot-action-recognition-on-hmdb51","task":"Zero-Shot Action Recognition","dataset":"HMDB51","model":"O2A","rank_in_archive_order":27,"of":29,"metrics":{"Top-1 Accuracy":"15.6"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-action-recognition-on-ucf101","task":"Zero-Shot Action Recognition","dataset":"UCF101","model":"O2A","rank_in_archive_order":24,"of":35,"metrics":{"Top-1 Accuracy":"30.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1510.06939","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}