{"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-regression-and-verification-networks","title":"Learning regression and verification networks for long-term visual tracking","arxiv_id":"1809.04320","date":"2018-09-12","proceeding":null,"authors":["Yunhua Zhang","Dong Wang","Lijun Wang","Jinqing Qi","Huchuan Lu"],"abstract":"Compared with short-term tracking, the long-term tracking task requires\ndetermining the tracked object is present or absent, and then estimating the\naccurate bounding box if present or conducting image-wide re-detection if\nabsent. Until now, few attempts have been done although this task is much\ncloser to designing practical tracking systems. In this work, we propose a\nnovel long-term tracking framework based on deep regression and verification\nnetworks. The offline-trained regression model is designed using the\nobject-aware feature fusion and region proposal networks to generate a series\nof candidates and estimate their similarity scores effectively. The\nverification network evaluates these candidates to output the optimal one as\nthe tracked object with its classification score, which is online updated to\nadapt to the appearance variations based on newly reliable observations. The\nsimilarity and classification scores are combined to obtain a final confidence\nvalue, based on which our tracker can determine the absence of the target\naccurately and conduct image-wide re-detection to capture the target\nsuccessfully when it reappears. Extensive experiments show that our tracker\nachieves the best performance on the VOT2018 long-term challenge and\nstate-of-the-art results on the OxUvA long-term dataset.","url_abs":"http://arxiv.org/abs/1809.04320v2","url_pdf":"http://arxiv.org/pdf/1809.04320v2.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-regression-and-verification-networks","repo_url":"https://github.com/hibetterheyj/Note-Everyday","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"learning-regression-and-verification-networks","repo_url":"https://github.com/hibetterheyj/Paper-Everyday","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"learning-regression-and-verification-networks","repo_url":"https://github.com/xiaobai1217/MBMD","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"object","task_name":"Object"},{"task_slug":"region-proposal","task_name":"Region Proposal"},{"task_slug":"visual-tracking","task_name":"Visual Tracking"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.04320","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}