{"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/enkcf-ensemble-of-kernelized-correlation","title":"EnKCF: Ensemble of Kernelized Correlation Filters for High-Speed Object Tracking","arxiv_id":"1801.06729","date":"2018-01-20","proceeding":null,"authors":["Burak Uzkent","Young-Woo Seo"],"abstract":"Computer vision technologies are very attractive for practical applications\nrunning on embedded systems. For such an application, it is desirable for the\ndeployed algorithms to run in high-speed and require no offline training. To\ndevelop a single-target tracking algorithm with these properties, we propose an\nensemble of the kernelized correlation filters (KCF), we call it EnKCF. A\ncommittee of KCFs is specifically designed to address the variations in scale\nand translation of moving objects. To guarantee a high-speed run-time\nperformance, we deploy each of KCFs in turn, instead of applying multiple KCFs\nto each frame. To minimize any potential drifts between individual KCFs\ntransition, we developed a particle filter. Experimental results showed that\nthe performance of ours is, on average, 70.10% for precision at 20 pixels,\n53.00% for success rate for the OTB100 data, and 54.50% and 40.2% for the\nUAV123 data. Experimental results showed that our method is better than other\nhigh-speed trackers over 5% on precision on 20 pixels and 10-20% on AUC on\naverage. Moreover, our implementation ran at 340 fps for the OTB100 and at 416\nfps for the UAV123 dataset that is faster than DCF (292 fps) for the OTB100 and\nKCF (292 fps) for the UAV123. To increase flexibility of the proposed EnKCF\nrunning on various platforms, we also explored different levels of deep\nconvolutional features.","url_abs":"http://arxiv.org/abs/1801.06729v1","url_pdf":"http://arxiv.org/pdf/1801.06729v1.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":"enkcf-ensemble-of-kernelized-correlation","repo_url":"https://github.com/buzkent86/EnKCF_Tracker","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"enkcf-ensemble-of-kernelized-correlation","repo_url":"https://github.com/buzkent86/EnKCF_Tracking_WACV18","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"high","task_name":"Vocal Bursts Intensity Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}