{"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-person-re-identification-with-improved","title":"Deep Person Re-Identification with Improved Embedding and Efficient Training","arxiv_id":"1705.03332","date":"2017-05-09","proceeding":null,"authors":["Haibo Jin","Xiaobo Wang","Shengcai Liao","Stan Z. Li"],"abstract":"Person re-identification task has been greatly boosted by deep convolutional\nneural networks (CNNs) in recent years. The core of which is to enlarge the\ninter-class distinction as well as reduce the intra-class variance. However, to\nachieve this, existing deep models prefer to adopt image pairs or triplets to\nform verification loss, which is inefficient and unstable since the number of\ntraining pairs or triplets grows rapidly as the number of training data grows.\nMoreover, their performance is limited since they ignore the fact that\ndifferent dimension of embedding may play different importance. In this paper,\nwe propose to employ identification loss with center loss to train a deep model\nfor person re-identification. The training process is efficient since it does\nnot require image pairs or triplets for training while the inter-class\ndistinction and intra-class variance are well handled. To boost the\nperformance, a new feature reweighting (FRW) layer is designed to explicitly\nemphasize the importance of each embedding dimension, thus leading to an\nimproved embedding. Experiments on several benchmark datasets have shown the\nsuperiority of our method over the state-of-the-art alternatives on both\naccuracy and speed.","url_abs":"http://arxiv.org/abs/1705.03332v3","url_pdf":"http://arxiv.org/pdf/1705.03332v3.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-person-re-identification-with-improved","repo_url":"https://github.com/jhb86253817/tf-re-id","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"person-re-identification","task_name":"Person Re-Identification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}