Reweighted Low-Rank Tensor Decomposition and its Applications in Video Denoising

A tensor is decomposed into low-rank and sparse components by simultaneously minimizing tensor nuclear norm and the $l_1$ norm in Tensor Principal Component Pursuit (TPCP). Inspired by reweighted $l_1$ norm minimization for sparsity enhancement, this paper proposes a reweighted singular value enhancement scheme to improve tensor low tubular rank in TPCP. The sparse component of a tensor is also recovered by the reweighted $l_1$ norm which enhances the accuracy of decomposition. An efficient iterative decomposition scheme based on t-SVD is proposed which improves low-rank signal recovery significantly. The effectiveness of the proposed method is established by applying to video denoising problem, and experimental results reveal that the algorithm outperforms its counterparts.

[1]  Yi Ma,et al.  Robust principal component analysis? , 2009, JACM.

[2]  Joos Vandewalle,et al.  A Multilinear Singular Value Decomposition , 2000, SIAM J. Matrix Anal. Appl..

[3]  J. Chang,et al.  Analysis of individual differences in multidimensional scaling via an n-way generalization of “Eckart-Young” decomposition , 1970 .

[4]  Eric L. Miller,et al.  Tensor-Based Formulation and Nuclear Norm Regularization for Multienergy Computed Tomography , 2013, IEEE Transactions on Image Processing.

[5]  Donald Goldfarb,et al.  Robust Low-Rank Tensor Recovery: Models and Algorithms , 2013, SIAM J. Matrix Anal. Appl..

[6]  Zemin Zhang,et al.  Exact Tensor Completion Using t-SVD , 2015, IEEE Transactions on Signal Processing.

[7]  Tamara G. Kolda,et al.  Tensor Decompositions and Applications , 2009, SIAM Rev..

[8]  Misha Elena Kilmer,et al.  Novel Methods for Multilinear Data Completion and De-noising Based on Tensor-SVD , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[9]  Liangpei Zhang,et al.  Total-Variation-Regularized Low-Rank Matrix Factorization for Hyperspectral Image Restoration , 2016, IEEE Transactions on Geoscience and Remote Sensing.

[10]  Yuan Xie,et al.  A New Low-Rank Tensor Model for Video Completion , 2015, ArXiv.

[11]  Stephen P. Boyd,et al.  Enhancing Sparsity by Reweighted ℓ1 Minimization , 2007, 0711.1612.

[12]  Pierre Comon,et al.  A Finite Algorithm to Compute Rank-1 Tensor Approximations , 2016, IEEE Signal Processing Letters.

[13]  Qionghai Dai,et al.  Reweighted Low-Rank Matrix Recovery and its Application in Image Restoration , 2014, IEEE Transactions on Cybernetics.

[14]  Hongyu Zhao,et al.  Low-Rank Modeling and Its Applications in Image Analysis , 2014, ACM Comput. Surv..

[15]  Carla D. Martin,et al.  An Order-p Tensor Factorization with Applications in Imaging , 2013, SIAM J. Sci. Comput..

[16]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..

[17]  Wei Liu,et al.  Tensor Robust Principal Component Analysis: Exact Recovery of Corrupted Low-Rank Tensors via Convex Optimization , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Misha Elena Kilmer,et al.  Third-Order Tensors as Operators on Matrices: A Theoretical and Computational Framework with Applications in Imaging , 2013, SIAM J. Matrix Anal. Appl..