{"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/person-re-identification-via-recurrent","title":"Person Re-Identification via Recurrent Feature Aggregation","arxiv_id":"1701.06351","date":"2017-01-23","proceeding":null,"authors":["Yichao Yan","Bingbing Ni","Zhichao Song","Chao Ma","Yan Yan","Xiaokang Yang"],"abstract":"We address the person re-identification problem by effectively exploiting a\nglobally discriminative feature representation from a sequence of tracked human\nregions/patches. This is in contrast to previous person re-id works, which rely\non either single frame based person to person patch matching, or graph based\nsequence to sequence matching. We show that a progressive/sequential fusion\nframework based on long short term memory (LSTM) network aggregates the\nframe-wise human region representation at each time stamp and yields a sequence\nlevel human feature representation. Since LSTM nodes can remember and propagate\npreviously accumulated good features and forget newly input inferior ones, even\nwith simple hand-crafted features, the proposed recurrent feature aggregation\nnetwork (RFA-Net) is effective in generating highly discriminative sequence\nlevel human representations. Extensive experimental results on two person\nre-identification benchmarks demonstrate that the proposed method performs\nfavorably against state-of-the-art person re-identification methods.","url_abs":"http://arxiv.org/abs/1701.06351v1","url_pdf":"http://arxiv.org/pdf/1701.06351v1.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":"person-re-identification-via-recurrent","repo_url":"https://github.com/daodaofr/caffe-re-id","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"patch-matching","task_name":"Patch Matching"},{"task_slug":"person-re-identification","task_name":"Person Re-Identification"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1701.06351","atlas_url":"https://app.syntology.ai/?focus=1701.06351","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}