{"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/scalable-person-re-identification-on","title":"Scalable Person Re-identification on Supervised Smoothed Manifold","arxiv_id":"1703.08359","date":"2017-03-24","proceeding":"CVPR 2017 7","authors":["Song Bai","Xiang Bai","Qi Tian"],"abstract":"Most existing person re-identification algorithms either extract robust\nvisual features or learn discriminative metrics for person images. However, the\nunderlying manifold which those images reside on is rarely investigated. That\nraises a problem that the learned metric is not smooth with respect to the\nlocal geometry structure of the data manifold.\n  In this paper, we study person re-identification with manifold-based affinity\nlearning, which did not receive enough attention from this area. An\nunconventional manifold-preserving algorithm is proposed, which can 1) make the\nbest use of supervision from training data, whose label information is given as\npairwise constraints; 2) scale up to large repositories with low on-line time\ncomplexity; and 3) be plunged into most existing algorithms, serving as a\ngeneric postprocessing procedure to further boost the identification\naccuracies. Extensive experimental results on five popular person\nre-identification benchmarks consistently demonstrate the effectiveness of our\nmethod. Especially, on the largest CUHK03 and Market-1501, our method\noutperforms the state-of-the-art alternatives by a large margin with high\nefficiency, which is more appropriate for practical applications.","url_abs":"http://arxiv.org/abs/1703.08359v1","url_pdf":"http://arxiv.org/pdf/1703.08359v1.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":"person-re-identification","task_name":"Person Re-Identification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/person-re-identification-on-market-1501","task":"Person Re-Identification","dataset":"Market-1501","model":"SSM","rank_in_archive_order":110,"of":135,"metrics":{"Rank-1":"82.21","mAP":"68.80"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.08359","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}