{"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/spatial-and-temporal-mutual-promotion-for","title":"Spatial and Temporal Mutual Promotion for Video-based Person Re-identification","arxiv_id":"1812.10305","date":"2018-12-26","proceeding":null,"authors":["Yiheng Liu","Zhenxun Yuan","Wengang Zhou","Houqiang Li"],"abstract":"Video-based person re-identification is a crucial task of matching video\nsequences of a person across multiple camera views. Generally, features\ndirectly extracted from a single frame suffer from occlusion, blur,\nillumination and posture changes. This leads to false activation or missing\nactivation in some regions, which corrupts the appearance and motion\nrepresentation. How to explore the abundant spatial-temporal information in\nvideo sequences is the key to solve this problem. To this end, we propose a\nRefining Recurrent Unit (RRU) that recovers the missing parts and suppresses\nnoisy parts of the current frame's features by referring historical frames.\nWith RRU, the quality of each frame's appearance representation is improved.\nThen we use the Spatial-Temporal clues Integration Module (STIM) to mine the\nspatial-temporal information from those upgraded features. Meanwhile, the\nmulti-level training objective is used to enhance the capability of RRU and\nSTIM. Through the cooperation of those modules, the spatial and temporal\nfeatures mutually promote each other and the final spatial-temporal feature\nrepresentation is more discriminative and robust. Extensive experiments are\nconducted on three challenging datasets, i.e., iLIDS-VID, PRID-2011 and MARS.\nThe experimental results demonstrate that our approach outperforms existing\nstate-of-the-art methods of video-based person re-identification on iLIDS-VID\nand MARS and achieves favorable results on PRID-2011.","url_abs":"http://arxiv.org/abs/1812.10305v1","url_pdf":"http://arxiv.org/pdf/1812.10305v1.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":"spatial-and-temporal-mutual-promotion-for","repo_url":"https://github.com/yolomax/rru-reid","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"spatial-and-temporal-mutual-promotion-for","repo_url":"https://github.com/MindSpore-scientific/code-14/tree/main/Spatial%20and%20Temporal%20Mutual%20Promotion%20for%20Video-based%20Person%20Re-identification","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"spatial-and-temporal-mutual-promotion-for","repo_url":"https://github.com/MindSpore-scientific/code-5/tree/main/Spatial%20and%20Temporal%20Mutual%20Promotion%20for%20Video-based%20Person%20Re-identification","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"spatial-and-temporal-mutual-promotion-for","repo_url":"https://github.com/mindspore-ai/contrib/tree/master/application/Spatial%20and%20Temporal%20Mutual%20Promotion%20for%20Video-based%20Person%20Re-identification","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null}],"tasks":[{"task_slug":"person-re-identification","task_name":"Person Re-Identification"},{"task_slug":"video-based-person-re-identification","task_name":"Video-Based Person Re-Identification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.10305","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}