Nonlinear Asymmetric Multi-Valued Hashing

doi:10.1109/TPAMI.2018.2867866

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 遥感图像处理团队

	Nonlinear Asymmetric Multi-Valued Hashing
	Da, Cheng1,2 ; Meng, Gaofeng1 ; Xiang, Shiming1,2 ; Ding, Kun1 ; Xu, Shibiao1 ; Yang, Qing1 ; Pan, Chunhong1
刊名	IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
	2019-11-01
卷号	41 期号:11 页码:2660-2676
关键词	Asymmetric hashing multi-valued embeddings binary sparse representation nonlinear transformation
ISSN号	0162-8828
DOI	10.1109/TPAMI.2018.2867866
通讯作者	Xiang, Shiming(smxiang@nlpr.ia.ac.cn)
英文摘要	Most existing hashing methods resort to binary codes for large scale similarity search, owing to the high efficiency of computation and storage. However, binary codes lack enough capability in similarity preservation, resulting in less desirable performance. To address this issue, we propose Nonlinear Asymmetric Multi-Valued Hashing (NAMVH) supported by two distinct non-binary embeddings. Specifically, a real-valued embedding is used for representing the newly-coming query by an ideally nonlinear transformation. Besides, a multi-integer-embedding is employed for compressing the whole database, which is modeled by Binary Sparse Representation (BSR) with fixed sparsity. With these two non-binary embeddings, NAMVH preserves more precise similarities between data points and enables access to the incremental extension with database samples evolving dynamically. To perform meaningful asymmetric similarity computation for efficient semantic search, these embeddings are jointly learnt by preserving the pairwise label-based similarity. Technically, this results in a mixed integer programming problem, which is efficiently solved by a well-designed alternative optimization method. Extensive experiments on seven large scale datasets demonstrate that our approach not only outperforms the existing binary hashing methods in search accuracy, but also retains their query and storage efficiency.
资助项目	National Natural Science Foundation of China[91646207] ; National Natural Science Foundation of China[61620106003] ; National Natural Science Foundation of China[61573352] ; National Natural Science Foundation of China[61671451]
WOS关键词	LEARNING BINARY-CODES ; RANKING ; OBJECT ; SCENE
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	IEEE COMPUTER SOC
WOS记录号	WOS:000489838200008
资助机构	National Natural Science Foundation of China
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/21706]
专题	自动化研究所_模式识别国家重点实验室_遥感图像处理团队自动化研究所_模式识别国家重点实验室自动化研究所_空天信息研究中心先进数据分析与学习团队遥感图像处理团队
通讯作者	Xiang, Shiming
作者单位	1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 101408, Peoples R China
推荐引用方式 GB/T 7714	Da, Cheng,Meng, Gaofeng,Xiang, Shiming,et al. Nonlinear Asymmetric Multi-Valued Hashing[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2019,41(11):2660-2676.
APA	Da, Cheng.,Meng, Gaofeng.,Xiang, Shiming.,Ding, Kun.,Xu, Shibiao.,...&Pan, Chunhong.(2019).Nonlinear Asymmetric Multi-Valued Hashing.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,41(11),2660-2676.
MLA	Da, Cheng,et al."Nonlinear Asymmetric Multi-Valued Hashing".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 41.11(2019):2660-2676.