Geometry-Aware Similarity Learning on SPD Manifolds for Visual Recognition

doi:10.1109/TCSVT.2017.2729660

	Geometry-Aware Similarity Learning on SPD Manifolds for Visual Recognition
	Huang, Zhiwu 1,2; Wang, Ruiping 1; Li, Xianqiu 1; Liu, Wenxian 1; Shan, Shiguang 1; Van Gool, Luc 2; Chen, Xilin 1
	2018-10-01
发表期刊	IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
ISSN	1051-8215
卷号	28 期号:10 页码:2513-2523
摘要	Symmetric positive definite (SPD) matrices have been employed for data representation in many visual recognition tasks. The success is mainly attributed to learning discriminative SPD matrices encoding the Riemannian geometry of the underlying SPD manifolds. In this paper, we propose a geometry-aware SPD similarity learning (SPDSL) framework to learn discriminative SPD features by directly pursuing a manifold-manifold transformation matrix of full column rank. Specifically, by exploiting the Riemannian geometry of the manifolds of fixed-rank positive semidefinite (PSD) matrices, we present a new solution to reduce optimization over the space of column full-rank transformation matrices to optimization on the PSD manifold, which has a well-established Riemannian structure. Under this solution, we exploit a new supervised SPDSL technique to learn the manifold-manifold transformation by regressing the similarities of selected SPD data pairs to their ground-truth similarities on the target SPD manifold. To optimize the proposed objective function, we further derive an optimization algorithm on the PSD manifold. Evaluations on three visual classification tasks show the advantages of the proposed approach over the existing SPD-based discriminant learning methods.
关键词	Discriminative SPD matrices Riemannian geometry SPD manifold geometry-aware SPD similarity learning PSD manifold
DOI	10.1109/TCSVT.2017.2729660
收录类别	SCI
语种	英语
资助项目	Natural Science Foundation of China[61390511] ; Natural Science Foundation of China[61379083] ; Natural Science Foundation of China[61650202] ; 973 Program[2015CB351802] ; Youth Innovation Promotion Association CAS[2015085]
WOS研究方向	Engineering
WOS类目	Engineering, Electrical & Electronic
WOS记录号	WOS:000448517900007
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：20[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/3642
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Wang, Ruiping
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 2.ETH, Swiss Fed Inst Technol, Comp Vis Lab, CH-8092 Zurich, Switzerland
推荐引用方式 GB/T 7714	Huang, Zhiwu,Wang, Ruiping,Li, Xianqiu,et al. Geometry-Aware Similarity Learning on SPD Manifolds for Visual Recognition[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,2018,28(10):2513-2523.
APA	Huang, Zhiwu.,Wang, Ruiping.,Li, Xianqiu.,Liu, Wenxian.,Shan, Shiguang.,...&Chen, Xilin.(2018).Geometry-Aware Similarity Learning on SPD Manifolds for Visual Recognition.IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,28(10),2513-2523.
MLA	Huang, Zhiwu,et al."Geometry-Aware Similarity Learning on SPD Manifolds for Visual Recognition".IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 28.10(2018):2513-2523.