{"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/fd-gan-pose-guided-feature-distilling-gan-for","title":"FD-GAN: Pose-guided Feature Distilling GAN for Robust Person Re-identification","arxiv_id":"1810.02936","date":"2018-10-06","proceeding":"NeurIPS 2018 12","authors":["Yixiao Ge","Zhuowan Li","Haiyu Zhao","Guojun Yin","Shuai Yi","Xiaogang Wang","Hongsheng Li"],"abstract":"Person re-identification (reID) is an important task that requires to\nretrieve a person's images from an image dataset, given one image of the person\nof interest. For learning robust person features, the pose variation of person\nimages is one of the key challenges. Existing works targeting the problem\neither perform human alignment, or learn human-region-based representations.\nExtra pose information and computational cost is generally required for\ninference. To solve this issue, a Feature Distilling Generative Adversarial\nNetwork (FD-GAN) is proposed for learning identity-related and pose-unrelated\nrepresentations. It is a novel framework based on a Siamese structure with\nmultiple novel discriminators on human poses and identities. In addition to the\ndiscriminators, a novel same-pose loss is also integrated, which requires\nappearance of a same person's generated images to be similar. After learning\npose-unrelated person features with pose guidance, no auxiliary pose\ninformation and additional computational cost is required during testing. Our\nproposed FD-GAN achieves state-of-the-art performance on three person reID\ndatasets, which demonstrates that the effectiveness and robust feature\ndistilling capability of the proposed FD-GAN.","url_abs":"http://arxiv.org/abs/1810.02936v2","url_pdf":"http://arxiv.org/pdf/1810.02936v2.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":"fd-gan-pose-guided-feature-distilling-gan-for","repo_url":"https://github.com/yxgeee/FD-GAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"fd-gan-pose-guided-feature-distilling-gan-for","repo_url":"https://github.com/NVlabs/DG-Net","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":null,"task_name":"Generative Adversarial Network"},{"task_slug":"person-re-identification","task_name":"Person Re-Identification"}],"methods":[{"method_slug":null,"method_name":null}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/person-re-identification-on-cuhk03","task":"Person Re-Identification","dataset":"CUHK03","model":"FD-GAN","rank_in_archive_order":3,"of":19,"metrics":{"MAP":"91.3","Rank-1":"92.6"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-dukemtmc-reid","task":"Person Re-Identification","dataset":"DukeMTMC-reID","model":"FD-GAN","rank_in_archive_order":71,"of":94,"metrics":{"Rank-1":"80.0","mAP":"64.5"},"uses_additional_data":false},{"leaderboard":"/sota/person-re-identification-on-market-1501","task":"Person Re-Identification","dataset":"Market-1501","model":"FD-GAN","rank_in_archive_order":92,"of":135,"metrics":{"Rank-1":"90.5","mAP":"77.7"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1810.02936","atlas_url":"https://app.syntology.ai/?focus=1810.02936","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}