{"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/deep-association-learning-for-unsupervised","title":"Deep Association Learning for Unsupervised Video Person Re-identification","arxiv_id":"1808.07301","date":"2018-08-22","proceeding":null,"authors":["Yanbei Chen","Xiatian Zhu","Shaogang Gong"],"abstract":"Deep learning methods have started to dominate the research progress of\nvideo-based person re-identification (re-id). However, existing methods mostly\nconsider supervised learning, which requires exhaustive manual efforts for\nlabelling cross-view pairwise data. Therefore, they severely lack scalability\nand practicality in real-world video surveillance applications. In this work,\nto address the video person re-id task, we formulate a novel Deep Association\nLearning (DAL) scheme, the first end-to-end deep learning method using none of\nthe identity labels in model initialisation and training. DAL learns a deep\nre-id matching model by jointly optimising two margin-based association losses\nin an end-to-end manner, which effectively constrains the association of each\nframe to the best-matched intra-camera representation and cross-camera\nrepresentation. Existing standard CNNs can be readily employed within our DAL\nscheme. Experiment results demonstrate that our proposed DAL significantly\noutperforms current state-of-the-art unsupervised video person re-id methods on\nthree benchmarks: PRID 2011, iLIDS-VID and MARS.","url_abs":"http://arxiv.org/abs/1808.07301v1","url_pdf":"http://arxiv.org/pdf/1808.07301v1.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":"deep-association-learning-for-unsupervised","repo_url":"https://github.com/yanbeic/Deep-Association-Learning","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"person-re-identification","task_name":"Person Re-Identification"},{"task_slug":"unsupervised-person-re-identification","task_name":"Unsupervised Person Re-Identification"},{"task_slug":"video-based-person-re-identification","task_name":"Video-Based Person Re-Identification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/person-re-identification-on-prid2011","task":"Person Re-Identification","dataset":"PRID2011","model":"DAL","rank_in_archive_order":6,"of":13,"metrics":{"Rank-1":"85.3","Rank-20":"99.6","Rank-5":"97.0"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}