{"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/learning-feature-fusion-for-unsupervised","title":"Learning Feature Fusion for Unsupervised Domain Adaptive Person Re-identification","arxiv_id":"2205.09495","date":"2022-05-19","proceeding":null,"authors":["Jin Ding","Xue Zhou"],"abstract":"Unsupervised domain adaptive (UDA) person re-identification (ReID) has gained increasing attention for its effectiveness on the target domain without manual annotations. Most fine-tuning based UDA person ReID methods focus on encoding global features for pseudo labels generation, neglecting the local feature that can provide for the fine-grained information. To handle this issue, we propose a Learning Feature Fusion (LF2) framework for adaptively learning to fuse global and local features to obtain a more comprehensive fusion feature representation. Specifically, we first pre-train our model within a source domain, then fine-tune the model on unlabeled target domain based on the teacher-student training strategy. The average weighting teacher network is designed to encode global features, while the student network updating at each iteration is responsible for fine-grained local features. By fusing these multi-view features, multi-level clustering is adopted to generate diverse pseudo labels. In particular, a learnable Fusion Module (FM) for giving prominence to fine-grained local information within the global feature is also proposed to avoid obscure learning of multiple pseudo labels. Experiments show that our proposed LF2 framework outperforms the state-of-the-art with 73.5% mAP and 83.7% Rank1 on Market1501 to DukeMTMC-ReID, and achieves 83.2% mAP and 92.8% Rank1 on DukeMTMC-ReID to Market1501.","url_abs":"https://arxiv.org/abs/2205.09495v1","url_pdf":"https://arxiv.org/pdf/2205.09495v1.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":"learning-feature-fusion-for-unsupervised","repo_url":"https://github.com/DJEddyking/LF2","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"unsupervised-domain-adaptation","task_name":"Unsupervised Domain Adaptation"},{"task_slug":null,"task_name":"Unsupervised Domain Adaptation on Duke to Market"},{"task_slug":null,"task_name":"Unsupervised Domain Adaptation on Market to Duke"},{"task_slug":"unsupervised-domain-adaptationn","task_name":"Unsupervised Domain Adaptationn"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/unsupervised-domain-adaptation-on-duke-to","task":"Unsupervised Domain Adaptation","dataset":"Duke to Market","model":"LF2","rank_in_archive_order":3,"of":26,"metrics":{"mAP":"83.2","rank-1":"92.8"},"uses_additional_data":false},{"leaderboard":"/sota/unsupervised-domain-adaptation-on-market-to","task":"Unsupervised Domain Adaptation","dataset":"Market to Duke","model":"LF2","rank_in_archive_order":2,"of":25,"metrics":{"mAP":"73.5","rank-1":"83.7"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}