Tensor RPCA by Bayesian CP Factorization with Complex Noise

CORC > 沈阳自动化研究所 > 中国科学院沈阳自动化研究所 > 机器人学研究室

	Tensor RPCA by Bayesian CP Factorization with Complex Noise
	Chen XA(陈希爱); Luo Q(罗琼); Liang D(梁栋); Meng DY(孟德宇); Wang Y(王尧); Han Z(韩志); Tang YD(唐延东)
	2017
会议名称	2017 IEEE International Conference on Computer Vision (ICCV)
会议日期	October 22-29, 2017
会议地点	Venice, Italy
页码	5029-5038
通讯作者	Han Z(韩志)
中文摘要	The RPCA model has achieved good performances in various applications. However, two defects limit its effectiveness. Firstly, it is designed for dealing with data in matrix form, which fails to exploit the structure information of higher order tensor data in some pratical situations. Secondly, it adopts L1-norm to tackle noise part which makes it only valid for sparse noise. In this paper, we propose a tensor RPCA model based on CP decomposition and model data noise by Mixture of Gaussians (MoG). The use of tensor structure to raw data allows us to make full use of the inherent structure priors, and MoG is a general approximator to any blends of consecutive distributions, which makes our approach capable of regaining the low dimensional linear subspace from a wide range of noises or their mixture. The model is solved by a new proposed algorithm inferred under a variational Bayesian framework. The superiority of our approach over the existing state-of-the-art approaches is demonstrated by extensive experiments on both of synthetic and real data.
收录类别	EI ; CPCI（ISTP）
产权排序	1
会议录	2017 IEEE International Conference on Computer Vision (ICCV)
会议录出版者	IEEE
会议录出版地	New York
语种	英语
ISSN号	2380-7504
ISBN号	978-1-5386-1032-9
WOS记录号	WOS:000425498405012
内容类型	会议论文
源URL	[http://ir.sia.cn/handle/173321/21357]
专题	沈阳自动化研究所_机器人学研究室
作者单位	1.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, China 2.Xi'An Jiaotong University, China 3.University of Chinese Academy of Sciences, China
推荐引用方式 GB/T 7714	Chen XA,Luo Q,Liang D,et al. Tensor RPCA by Bayesian CP Factorization with Complex Noise[C]. 见:2017 IEEE International Conference on Computer Vision (ICCV). Venice, Italy. October 22-29, 2017.