Accurate and efficient cross-domain visual matching leveraging multiple feature representations

ISCAS OpenIR

	Accurate and efficient cross-domain visual matching leveraging multiple feature representations
	Sun, Gang (1); Wang, Shuhui (3); Liu, Xuehui (1); Huang, Qingming (2); Chen, Yanyun (1); Wu, Enhua (1); Sun, G.(sung@ios.ac.cn)
	2013
页码	565-575
收录类别	SCI ; EI
出版地	Springer Verlag
ISSN	1782789
部门归属	(1) State Key Laboratory of Computer Science, Institute of Software, Chinese Academy of Sciences, Beijing, China; (2) University of Chinese Academy of Sciences, Beijing, China; (3) Key Laboratory of Intelligent Information Processing (CAS), Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China; (4) University of Macau, China
摘要	Cross-domain visual matching aims at finding visually similar images across a wide range of visual domains, and has shown a practical impact on a number of applications. Unfortunately, the state-of-the-art approach, which estimates the relative importance of the single feature dimensions still suffers from low matching accuracy and high time cost. To this end, this paper proposes a novel cross-domain visual matching framework leveraging multiple feature representations. To integrate the discriminative power of multiple features, we develop a data-driven, query specific feature fusion model, which estimates the relative importance of the individual feature dimensions as well as the weight vector among multiple features simultaneously. Moreover, to alleviate the computational burden of an exhaustive subimage search, we design a speedup scheme, which employs hyperplane hashing for rapidly collecting the hard-negatives. Extensive experiments carried out on various matching tasks demonstrate that the proposed approach outperforms the state-of-the-art in both accuracy and efficiency. © 2013 Springer-Verlag Berlin Heidelberg.; Cross-domain visual matching aims at finding visually similar images across a wide range of visual domains, and has shown a practical impact on a number of applications. Unfortunately, the state-of-the-art approach, which estimates the relative importance of the single feature dimensions still suffers from low matching accuracy and high time cost. To this end, this paper proposes a novel cross-domain visual matching framework leveraging multiple feature representations. To integrate the discriminative power of multiple features, we develop a data-driven, query specific feature fusion model, which estimates the relative importance of the individual feature dimensions as well as the weight vector among multiple features simultaneously. Moreover, to alleviate the computational burden of an exhaustive subimage search, we design a speedup scheme, which employs hyperplane hashing for rapidly collecting the hard-negatives. Extensive experiments carried out on various matching tasks demonstrate that the proposed approach outperforms the state-of-the-art in both accuracy and efficiency. © 2013 Springer-Verlag Berlin Heidelberg.
关键词	Visual Matching Cross-domain Multiple Features Hyperplane Hashing
语种	英语
WOS记录号	WOS:000319478400011
引用统计	被引频次：2[WOS] [WOS记录] [WOS相关记录]
内容类型	会议论文
URI标识	http://ir.iscas.ac.cn/handle/311060/16556
专题	中国科学院软件研究所
通讯作者	Sun, G.(sung@ios.ac.cn)
推荐引用方式 GB/T 7714	Sun, Gang ,Wang, Shuhui ,Liu, Xuehui ,et al. Accurate and efficient cross-domain visual matching leveraging multiple feature representations[C]. Springer Verlag,2013:565-575.