{"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-temporal-regularized","title":"Learning Spatial-Temporal Regularized Correlation Filters for Visual Tracking","arxiv_id":"1803.08679","date":"2018-03-23","proceeding":"CVPR 2018 6","authors":["Feng Li","Cheng Tian","WangMeng Zuo","Lei Zhang","Ming-Hsuan Yang"],"abstract":"Discriminative Correlation Filters (DCF) are efficient in visual tracking but\nsuffer from unwanted boundary effects. Spatially Regularized DCF (SRDCF) has\nbeen suggested to resolve this issue by enforcing spatial penalty on DCF\ncoefficients, which, inevitably, improves the tracking performance at the price\nof increasing complexity. To tackle online updating, SRDCF formulates its model\non multiple training images, further adding difficulties in improving\nefficiency. In this work, by introducing temporal regularization to SRDCF with\nsingle sample, we present our spatial-temporal regularized correlation filters\n(STRCF). Motivated by online Passive-Agressive (PA) algorithm, we introduce the\ntemporal regularization to SRDCF with single sample, thus resulting in our\nspatial-temporal regularized correlation filters (STRCF). The STRCF formulation\ncan not only serve as a reasonable approximation to SRDCF with multiple\ntraining samples, but also provide a more robust appearance model than SRDCF in\nthe case of large appearance variations. Besides, it can be efficiently solved\nvia the alternating direction method of multipliers (ADMM). By incorporating\nboth temporal and spatial regularization, our STRCF can handle boundary effects\nwithout much loss in efficiency and achieve superior performance over SRDCF in\nterms of accuracy and speed. Experiments are conducted on three benchmark\ndatasets: OTB-2015, Temple-Color, and VOT-2016. Compared with SRDCF, STRCF with\nhand-crafted features provides a 5 times speedup and achieves a gain of 5.4%\nand 3.6% AUC score on OTB-2015 and Temple-Color, respectively. Moreover, STRCF\ncombined with CNN features also performs favorably against state-of-the-art\nCNN-based trackers and achieves an AUC score of 68.3% on OTB-2015.","url_abs":"http://arxiv.org/abs/1803.08679v1","url_pdf":"http://arxiv.org/pdf/1803.08679v1.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-temporal-regularized","repo_url":"https://github.com/lifeng9472/STRCF","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"video-object-tracking","task_name":"Video Object Tracking"},{"task_slug":"visual-object-tracking","task_name":"Visual Object Tracking"},{"task_slug":"visual-tracking","task_name":"Visual Tracking"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/video-object-tracking-on-nv-vot211","task":"Video Object Tracking","dataset":"NT-VOT211","model":"STRCF","rank_in_archive_order":35,"of":43,"metrics":{"AUC":"27.86","Precision":"36.18"},"uses_additional_data":false},{"leaderboard":"/sota/visual-object-tracking-on-vot201718","task":"Visual Object Tracking","dataset":"VOT2017/18","model":"STRCF","rank_in_archive_order":9,"of":15,"metrics":{"Expected Average Overlap (EAO)":"0.345"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1803.08679","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}