{"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/alternative-semantic-representations-for-zero","title":"Alternative Semantic Representations for Zero-Shot Human Action Recognition","arxiv_id":"1706.09317","date":"2017-06-28","proceeding":null,"authors":["Qian Wang","Ke Chen"],"abstract":"A proper semantic representation for encoding side information is key to the\nsuccess of zero-shot learning. In this paper, we explore two alternative\nsemantic representations especially for zero-shot human action recognition:\ntextual descriptions of human actions and deep features extracted from still\nimages relevant to human actions. Such side information are accessible on Web\nwith little cost, which paves a new way in gaining side information for\nlarge-scale zero-shot human action recognition. We investigate different\nencoding methods to generate semantic representations for human actions from\nsuch side information. Based on our zero-shot visual recognition method, we\nconducted experiments on UCF101 and HMDB51 to evaluate two proposed semantic\nrepresentations . The results suggest that our proposed text- and image-based\nsemantic representations outperform traditional attributes and word vectors\nconsiderably for zero-shot human action recognition. In particular, the\nimage-based semantic representations yield the favourable performance even\nthough the representation is extracted from a small number of images per class.","url_abs":"http://arxiv.org/abs/1706.09317v1","url_pdf":"http://arxiv.org/pdf/1706.09317v1.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":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":"action-recognition","task_name":"Temporal Action Localization"},{"task_slug":"zero-shot-action-recognition","task_name":"Zero-Shot Action Recognition"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/zero-shot-action-recognition-on-hmdb51","task":"Zero-Shot Action Recognition","dataset":"HMDB51","model":"ASR","rank_in_archive_order":24,"of":29,"metrics":{"Top-1 Accuracy":"21.8"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-action-recognition-on-ucf101","task":"Zero-Shot Action Recognition","dataset":"UCF101","model":"ASR","rank_in_archive_order":25,"of":35,"metrics":{"Top-1 Accuracy":"24.4"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.09317","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}