{"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/leveraging-virtual-and-real-person-for","title":"Leveraging Virtual and Real Person for Unsupervised Person Re-identification","arxiv_id":"1811.02074","date":"2018-11-05","proceeding":null,"authors":["Fengxiang Yang","Zhun Zhong","Zhiming Luo","Sheng Lian","Shaozi Li"],"abstract":"Person re-identification (re-ID) is a challenging problem especially when no\nlabels are available for training. Although recent deep re-ID methods have\nachieved great improvement, it is still difficult to optimize deep re-ID model\nwithout annotations in training data. To address this problem, this study\nintroduces a novel approach for unsupervised person re-ID by leveraging virtual\nand real data. Our approach includes two components: virtual person generation\nand training of deep re-ID model. For virtual person generation, we learn a\nperson generation model and a camera style transfer model using unlabeled real\ndata to generate virtual persons with different poses and camera styles. The\nvirtual data is formed as labeled training data, enabling subsequently training\ndeep re-ID model in supervision. For training of deep re-ID model, we divide it\ninto three steps: 1) pre-training a coarse re-ID model by using virtual data;\n2) collaborative filtering based positive pair mining from the real data; and\n3) fine-tuning of the coarse re-ID model by leveraging the mined positive pairs\nand virtual data. The final re-ID model is achieved by iterating between step 2\nand step 3 until convergence. Experimental results on two large-scale datasets,\nMarket-1501 and DukeMTMC-reID, demonstrate the effectiveness of our approach\nand shows that the state of the art is achieved in unsupervised person re-ID.","url_abs":"http://arxiv.org/abs/1811.02074v1","url_pdf":"http://arxiv.org/pdf/1811.02074v1.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":"leveraging-virtual-and-real-person-for","repo_url":"https://github.com/FlyingRoastDuck/PGPPM","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"collaborative-filtering","task_name":"Collaborative Filtering"},{"task_slug":"person-re-identification","task_name":"Person Re-Identification"},{"task_slug":"style-transfer","task_name":"Style Transfer"},{"task_slug":"unsupervised-person-re-identification","task_name":"Unsupervised Person Re-Identification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.02074","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}