{"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/unsupervised-tracklet-person-re","title":"Unsupervised Tracklet Person Re-Identification","arxiv_id":"1903.00535","date":"2019-03-01","proceeding":null,"authors":["Minxian Li","Xiatian Zhu","Shaogang Gong"],"abstract":"Most existing person re-identification (re-id) methods rely on supervised\nmodel learning on per-camera-pair manually labelled pairwise training data.\nThis leads to poor scalability in a practical re-id deployment, due to the lack\nof exhaustive identity labelling of positive and negative image pairs for every\ncamera-pair. In this work, we present an unsupervised re-id deep learning\napproach. It is capable of incrementally discovering and exploiting the\nunderlying re-id discriminative information from automatically generated person\ntracklet data end-to-end. We formulate an Unsupervised Tracklet Association\nLearning (UTAL) framework. This is by jointly learning within-camera tracklet\ndiscrimination and cross-camera tracklet association in order to maximise the\ndiscovery of tracklet identity matching both within and across camera views.\nExtensive experiments demonstrate the superiority of the proposed model over\nthe state-of-the-art unsupervised learning and domain adaptation person re-id\nmethods on eight benchmarking datasets.","url_abs":"http://arxiv.org/abs/1903.00535v1","url_pdf":"http://arxiv.org/pdf/1903.00535v1.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":"unsupervised-tracklet-person-re","repo_url":"https://github.com/liminxian/DukeMTMC-SI-Tracklet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"benchmarking","task_name":"Benchmarking"},{"task_slug":"domain-adaptation","task_name":"Domain Adaptation"},{"task_slug":"person-re-identification","task_name":"Person Re-Identification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/person-re-identification-on-cuhk03","task":"Person Re-Identification","dataset":"CUHK03","model":"UTAL","rank_in_archive_order":14,"of":19,"metrics":{"MAP":"42.3","Rank-1":"56.3"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-dukemtmc-reid","task":"Person Re-Identification","dataset":"DukeMTMC-reID","model":"UTAL","rank_in_archive_order":88,"of":94,"metrics":{"Rank-1":"62.3","mAP":"44.6"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-duketracklet","task":"Person Re-Identification","dataset":"DukeTracklet","model":"UTAL","rank_in_archive_order":1,"of":2,"metrics":{"Rank-1":"43.8","Rank-20":"76.5","Rank-5":"62.8","mAP":"36.6"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-mars","task":"Person Re-Identification","dataset":"MARS","model":"UTAL","rank_in_archive_order":20,"of":21,"metrics":{"Rank-1":"49.9","Rank-10":"66.4","Rank-20":"77.8","mAP":"35.2"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-msmt17","task":"Person Re-Identification","dataset":"MSMT17","model":"UTAL","rank_in_archive_order":41,"of":43,"metrics":{"Rank-1":"31.4","mAP":"13.1"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-market-1501","task":"Person Re-Identification","dataset":"Market-1501","model":"UTAL","rank_in_archive_order":117,"of":135,"metrics":{"Rank-1":"69.2","mAP":"46.2"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-prid2011","task":"Person Re-Identification","dataset":"PRID2011","model":"UTAL","rank_in_archive_order":10,"of":13,"metrics":{"Rank-1":"54.7","Rank-20":"96.2","Rank-5":"83.1"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-ilids-vid","task":"Person Re-Identification","dataset":"iLIDS-VID","model":"UTAL","rank_in_archive_order":10,"of":10,"metrics":{"Rank-1":"35.1","Rank-20":"83.8","Rank-5":"59"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.00535","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}