{"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/text-based-person-search-via-attribute-aided","title":"Text-based Person Search via Attribute-aided Matching","arxiv_id":null,"date":"2020-03-14","proceeding":null,"authors":["Surbhi Aggarwal R.","Venkatesh Babu","Anirban Chakraborty"],"abstract":"Text-based person search aims to retrieve the pedestrian images that best match a given text query. Existing\r\nmethods utilize class-id information to get discriminative\r\nand identity-preserving features. However, it is not wellexplored whether it is beneficial to explicitly ensure that the\r\nsemantics of the data are retained. In the proposed work, we\r\naim to create semantics-preserving embeddings through an\r\nadditional task of attribute prediction. Since attribute annotation is typically unavailable in text-based person search,\r\nwe first mine them from the text corpus. These attributes are\r\nthen used as a means to bridge the modality gap between the\r\nimage-text inputs, as well as to improve the representation\r\nlearning. In summary, we propose an approach for textbased person search by learning an attribute-driven space\r\nalong with a class-information driven space, and utilize\r\nboth for obtaining the retrieval results. Our experiments on\r\nbenchmark dataset, CUHK-PEDES, show that learning the\r\nattribute-space not only helps in improving performance,\r\ngiving us state-of-the-art Rank-1 accuracy of 56.68%, but\r\nalso yields humanly-interpretable features.","url_abs":"https://openaccess.thecvf.com/content_WACV_2020/html/Aggarwal_Text-based_Person_Search_via_Attribute-aided_Matching_WACV_2020_paper.html","url_pdf":"https://openaccess.thecvf.com/content_WACV_2020/papers/Aggarwal_Text-based_Person_Search_via_Attribute-aided_Matching_WACV_2020_paper.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":"person-search","task_name":"Person Search"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"nlp-based-person-retrival","task_name":"Text based Person Retrieval"},{"task_slug":"text-based-person-search","task_name":"Text based Person Search"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/nlp-based-person-retrival-on-cuhk-pedes","task":"Text based Person Retrieval","dataset":"CUHK-PEDES","model":"CMAAM","rank_in_archive_order":15,"of":21,"metrics":{"R@1":"56.68","R@10":"84.86","R@5":"77.18"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}