QQ客服 官方微博 反馈留言

	Collaborative Correlation Tracking
	Zhu, Guibo1; Wang, Jinqiao1; Wu, Yi2; Lu, Hanqing1
	2015
会议日期	September 7-10
会议地点	Swansea, UK
关键词	Visual Tracking Collaborative Correlation Tracking
英文摘要	Correlation filter based tracking has attracted many researchers’ attention in recent years for high efficiency and robustness. Most existing works focus on exploiting different characteristics with correlation filters for visual tracking, e.g. circulant structure, kernel trick, effective feature representation and context information. However, how to handle the scale variation and the model drift is still an open problem. In this paper, we propose a collaborative correlation tracker to deal with the above problems. Firstly, we extend the correlation tracking filter by embedding the scale factor into the kernelized matrix to handle the scale variation. Then a novel long-term CUR filter for detection is learnt efficiently with random sampling to alleviate model drift by detecting effective object candidates in the collaborative tracker. In this way, the proposed approach could estimate the object state accurately and handle the model drift problem effectively. Extensive experiments show the superiority of the proposed method.
会议录	In proceedings of British Machine Computer Vision
内容类型	会议论文
源URL	[http://ir.ia.ac.cn/handle/173211/11756]
专题	自动化研究所_模式识别国家重点实验室_图像与视频分析团队
通讯作者	Wang, Jinqiao
作者单位	1.Chinese Acad Sci, Natl Lab Pattern Recognit, Inst Automat, Beijing 100190, Peoples R China 2.B-DAT & CICAEET, School of Information & Control, Nanjing University of Information Science and Technology, Nanjing, 210044, Jiangsu, China
推荐引用方式 GB/T 7714	Zhu, Guibo,Wang, Jinqiao,Wu, Yi,et al. Collaborative Correlation Tracking[C]. 见:. Swansea, UK. September 7-10.

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。