On Unifying Multi-view Self-Representations for Clustering by Tensor Multi-rank Minimization

doi:10.1007/s11263-018-1086-2

	On Unifying Multi-view Self-Representations for Clustering by Tensor Multi-rank Minimization
	Xie, Yuan2 ; Tao, Dacheng 3; Zhang, Wensheng2 ; Liu, Yan 1; Zhang, Lei 1; Qu, Yanyun 4
刊名	INTERNATIONAL JOURNAL OF COMPUTER VISION
	2018-11-01
卷号	126 期号:11 页码:1157-1179
关键词	T-svd Tensor Multi-rank Multi-view Features Subspace Clustering
ISSN号	0920-5691
DOI	10.1007/s11263-018-1086-2
文献子类	Article
英文摘要	In this paper, we address the multi-view subspace clustering problem. Our method utilizes the circulant algebra for tensor, which is constructed by stacking the subspace representation matrices of different views and then rotating, to capture the low rank tensor subspace so that the refinement of the view-specific subspaces can be achieved, as well as the high order correlations underlying multi-view data can be explored. By introducing a recently proposed tensor factorization, namely tensor-Singular Value Decomposition (t-SVD) (Kilmer et al. in SIAM J Matrix Anal Appl 34(1):148-172, 2013), we can impose a new type of low-rank tensor constraint on the rotated tensor to ensure the consensus among multiple views. Different from traditional unfolding based tensor norm, this low-rank tensor constraint has optimality properties similar to that of matrix rank derived from SVD, so the complementary information can be explored and propagated among all the views more thoroughly and effectively. The established model, called t-SVD based Multi-view Subspace Clustering (t-SVD-MSC), falls into the applicable scope of augmented Lagrangian method, and its minimization problem can be efficiently solved with theoretical convergence guarantee and relatively low computational complexity. Extensive experimental testing on eight challenging image datasets shows that the proposed method has achieved highly competent objective performance compared to several state-of-the-art multi-view clustering methods.
WOS关键词	3RD-ORDER TENSORS ; CATEGORIES ; ALGORITHM ; OPERATORS ; SCENE
WOS研究方向	Computer Science
语种	英语
出版者	SPRINGER
WOS记录号	WOS:000444394200001
资助机构	National Natural Science Foundation of China(61772524 ; Beijing Natural Science Foundation(4182067) ; Australian Research Council(FL-170100117 ; HK RGC General Research Fund(PolyU 152135/16E) ; 61402480 ; DP-180103424 ; 61373077 ; DP-140102164 ; 61602482) ; LP-150100671)
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/27918]
专题	中国科学院自动化研究所
通讯作者	Xie, Yuan
作者单位	1.Hong Kong Polytech Univ, Dept Comp, Hong Kong, Hong Kong, Peoples R China 2.Chinese Acad Sci, Inst Automat, Res Ctr Precis Sensing & Control, Beijing 100190, Peoples R China 3.Univ Sydney, Fac Engn & Informat Technol, Sch Informat Technol, UBTECH Sydney Artificial Intelligence Ctr, 6 Cleveland St, Darlington, NSW 2008, Australia 4.Xiamen Univ, Sch Informat Sci & Technol, Xiamen, Fujian, Peoples R China
推荐引用方式 GB/T 7714	Xie, Yuan,Tao, Dacheng,Zhang, Wensheng,et al. On Unifying Multi-view Self-Representations for Clustering by Tensor Multi-rank Minimization[J]. INTERNATIONAL JOURNAL OF COMPUTER VISION,2018,126(11):1157-1179.
APA	Xie, Yuan,Tao, Dacheng,Zhang, Wensheng,Liu, Yan,Zhang, Lei,&Qu, Yanyun.(2018).On Unifying Multi-view Self-Representations for Clustering by Tensor Multi-rank Minimization.INTERNATIONAL JOURNAL OF COMPUTER VISION,126(11),1157-1179.
MLA	Xie, Yuan,et al."On Unifying Multi-view Self-Representations for Clustering by Tensor Multi-rank Minimization".INTERNATIONAL JOURNAL OF COMPUTER VISION 126.11(2018):1157-1179.