On-line distance metric learning tracking using compressed feature

	On-line distance metric learning tracking using compressed feature
	Chen D.-C.; Zhu M.; He B.-G.; Yang W.-B.
	2014
发表期刊	Guangdianzi Jiguang/Journal of Optoelectronics Laser
ISSN	ISBN/10050086
卷号	25 期号:8 页码:1592-1597
摘要	To improve the precision and real-time quality of on-line learning object tracking, combined with the compress sensing theory, an algorithm using distance metric learning is proposed. First, target samples and background samples around the selected target are sampled. The Harr-like feature vectors are compressed using the random projection theory. Then, the distance metric is trained using the compressed feature vectors. Finally, the Mahalanobis distance between the samples in the new coming frame and the known target is calculated. The location of the sample closest to the known target is the location wanted. Experiments on variant videos show that the calculating load of the compressed features is 3/4 less than that using the uncompressed ones. Calculating the location of target using the trained distance metric makes the tracking precision higher.
收录类别	EI
语种	中文
文献类型	期刊论文
条目标识符	http://ir.ciomp.ac.cn/handle/181722/44207
专题	中科院长春光机所知识产出
推荐引用方式 GB/T 7714	Chen D.-C.,Zhu M.,He B.-G.,et al. On-line distance metric learning tracking using compressed feature[J]. Guangdianzi Jiguang/Journal of Optoelectronics Laser,2014,25(8):1592-1597.
APA	Chen D.-C.,Zhu M.,He B.-G.,&Yang W.-B..(2014).On-line distance metric learning tracking using compressed feature.Guangdianzi Jiguang/Journal of Optoelectronics Laser,25(8),1592-1597.
MLA	Chen D.-C.,et al."On-line distance metric learning tracking using compressed feature".Guangdianzi Jiguang/Journal of Optoelectronics Laser 25.8(2014):1592-1597.

条目包含的文件		下载所有文件
文件名称/大小	文献类型	版本类型	开放类型	使用许可
特征压缩在线距离度量学习跟踪_陈东成.c（430KB）			开放获取	CC BY-ND	浏览下载