{"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/svdnet-for-pedestrian-retrieval","title":"SVDNet for Pedestrian Retrieval","arxiv_id":"1703.05693","date":"2017-03-16","proceeding":"ICCV 2017 10","authors":["Yifan Sun","Liang Zheng","Weijian Deng","Shengjin Wang"],"abstract":"This paper proposes the SVDNet for retrieval problems, with focus on the\napplication of person re-identification (re-ID). We view each weight vector\nwithin a fully connected (FC) layer in a convolutional neuron network (CNN) as\na projection basis. It is observed that the weight vectors are usually highly\ncorrelated. This problem leads to correlations among entries of the FC\ndescriptor, and compromises the retrieval performance based on the Euclidean\ndistance. To address the problem, this paper proposes to optimize the deep\nrepresentation learning process with Singular Vector Decomposition (SVD).\nSpecifically, with the restraint and relaxation iteration (RRI) training\nscheme, we are able to iteratively integrate the orthogonality constraint in\nCNN training, yielding the so-called SVDNet. We conduct experiments on the\nMarket-1501, CUHK03, and Duke datasets, and show that RRI effectively reduces\nthe correlation among the projection vectors, produces more discriminative FC\ndescriptors, and significantly improves the re-ID accuracy. On the Market-1501\ndataset, for instance, rank-1 accuracy is improved from 55.3% to 80.5% for\nCaffeNet, and from 73.8% to 82.3% for ResNet-50.","url_abs":"http://arxiv.org/abs/1703.05693v4","url_pdf":"http://arxiv.org/pdf/1703.05693v4.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"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"retrieval","task_name":"Retrieval"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"bottleneck-residual-block","method_name":"Bottleneck Residual Block"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/person-re-identification-on-cuhk03-detected","task":"Person Re-Identification","dataset":"CUHK03 detected","model":"SVDNet-ResNet50","rank_in_archive_order":14,"of":19,"metrics":{"MAP":"37.3","Rank-1":"41.5"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-cuhk03-detected","task":"Person Re-Identification","dataset":"CUHK03 detected","model":"SVDNet-CaffeNet","rank_in_archive_order":16,"of":19,"metrics":{"MAP":"24.9","Rank-1":"27.7"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-dukemtmc-reid","task":"Person Re-Identification","dataset":"DukeMTMC-reID","model":"SVDNet","rank_in_archive_order":76,"of":94,"metrics":{"Rank-1":"76.7","mAP":"56.8"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-market-1501","task":"Person Re-Identification","dataset":"Market-1501","model":"SVDNet","rank_in_archive_order":109,"of":135,"metrics":{"Rank-1":"82.3","mAP":"62.1"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1703.05693","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}