{"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/parameter-free-spatial-attention-network-for","title":"Parameter-Free Spatial Attention Network for Person Re-Identification","arxiv_id":"1811.12150","date":"2018-11-29","proceeding":null,"authors":["Haoran Wang","Yue Fan","Zexin Wang","Licheng Jiao","Bernt Schiele"],"abstract":"Global average pooling (GAP) allows to localize discriminative information\nfor recognition [40]. While GAP helps the convolution neural network to attend\nto the most discriminative features of an object, it may suffer if that\ninformation is missing e.g. due to camera viewpoint changes. To circumvent this\nissue, we argue that it is advantageous to attend to the global configuration\nof the object by modeling spatial relations among high-level features. We\npropose a novel architecture for Person Re-Identification, based on a novel\nparameter-free spatial attention layer introducing spatial relations among the\nfeature map activations back to the model. Our spatial attention layer\nconsistently improves the performance over the model without it. Results on\nfour benchmarks demonstrate a superiority of our model over the\nstate-of-the-art achieving rank-1 accuracy of 94.7% on Market-1501, 89.0% on\nDukeMTMC-ReID, 74.9% on CUHK03-labeled and 69.7% on CUHK03-detected.","url_abs":"http://arxiv.org/abs/1811.12150v1","url_pdf":"http://arxiv.org/pdf/1811.12150v1.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":"parameter-free-spatial-attention-network-for","repo_url":"https://github.com/HRanWang/SA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"parameter-free-spatial-attention-network-for","repo_url":"https://github.com/HRanWang/Spatial-Attention","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"parameter-free-spatial-attention-network-for","repo_url":"https://github.com/schizop/SA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"person-re-identification","task_name":"Person Re-Identification"}],"methods":[{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/person-re-identification-on-dukemtmc-reid","task":"Person Re-Identification","dataset":"DukeMTMC-reID","model":"Parameter-Free Spatial Attention","rank_in_archive_order":20,"of":94,"metrics":{"Rank-1":"89.0","mAP":"85.9"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-market-1501","task":"Person Re-Identification","dataset":"Market-1501","model":"Parameter-Free Spatial Attention (RK)","rank_in_archive_order":76,"of":135,"metrics":{"Rank-1":"94.7","mAP":"91.7"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1811.12150","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}