{"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/sparse-label-smoothing-regularization-for","title":"Sparse Label Smoothing Regularization for Person Re-Identification","arxiv_id":"1809.04976","date":"2018-09-13","proceeding":null,"authors":["Jean-Paul Ainam","Ke Qin","Guisong Liu","Guangchun Luo"],"abstract":"Person re-identification (re-id) is a cross-camera retrieval task which\nestablishes a correspondence between images of a person from multiple cameras.\nDeep Learning methods have been successfully applied to this problem and have\nachieved impressive results. However, these methods require a large amount of\nlabeled training data. Currently labeled datasets in person re-id are limited\nin their scale and manual acquisition of such large-scale datasets from\nsurveillance cameras is a tedious and labor-intensive task. In this paper, we\npropose a framework that performs intelligent data augmentation and assigns\npartial smoothing label to generated data. Our approach first exploits the\nclustering property of existing person re-id datasets to create groups of\nsimilar objects that model cross-view variations. Each group is then used to\ngenerate realistic images through adversarial training. Our aim is to emphasize\nfeature similarity between generated samples and the original samples. Finally,\nwe assign a non-uniform label distribution to the generated samples and define\na regularized loss function for training. The proposed approach tackles two\nproblems (1) how to efficiently use the generated data and (2) how to address\nthe over-smoothness problem found in current regularization methods. Extensive\nexperiments on four larges cale datasets show that our regularization method\nsignificantly improves the Re-ID accuracy compared to existing methods.","url_abs":"http://arxiv.org/abs/1809.04976v3","url_pdf":"http://arxiv.org/pdf/1809.04976v3.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":"sparse-label-smoothing-regularization-for","repo_url":"https://github.com/jpainam/SLS_ReID","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"person-re-identification","task_name":"Person Re-Identification"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"semi-supervised-person-re-identification","task_name":"Semi-Supervised 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}