{"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-spatial-aware-regressions-for-visual","title":"Learning Spatial-Aware Regressions for Visual Tracking","arxiv_id":"1706.07457","date":"2017-06-22","proceeding":"CVPR 2018 6","authors":["Chong Sun","Dong Wang","Huchuan Lu","Ming-Hsuan Yang"],"abstract":"In this paper, we analyze the spatial information of deep features, and\npropose two complementary regressions for robust visual tracking. First, we\npropose a kernelized ridge regression model wherein the kernel value is defined\nas the weighted sum of similarity scores of all pairs of patches between two\nsamples. We show that this model can be formulated as a neural network and thus\ncan be efficiently solved. Second, we propose a fully convolutional neural\nnetwork with spatially regularized kernels, through which the filter kernel\ncorresponding to each output channel is forced to focus on a specific region of\nthe target. Distance transform pooling is further exploited to determine the\neffectiveness of each output channel of the convolution layer. The outputs from\nthe kernelized ridge regression model and the fully convolutional neural\nnetwork are combined to obtain the ultimate response. Experimental results on\ntwo benchmark datasets validate the effectiveness of the proposed method.","url_abs":"http://arxiv.org/abs/1706.07457v2","url_pdf":"http://arxiv.org/pdf/1706.07457v2.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-spatial-aware-regressions-for-visual","repo_url":"https://github.com/cswaynecool/LSART","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"visual-object-tracking","task_name":"Visual Object Tracking"},{"task_slug":"visual-tracking","task_name":"Visual Tracking"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/visual-object-tracking-on-vot201718","task":"Visual Object Tracking","dataset":"VOT2017/18","model":"LSART","rank_in_archive_order":12,"of":15,"metrics":{"Expected Average Overlap (EAO)":"0.323"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}