Novel methods for multilinear data completion and de-noising based on tensor-SVD

Abstract

In this paper we propose novel methods for completion (from limited samples) and de-noising of multilinear (tensor) data and as an application consider 3-D and 4- D (color) video data completion and de-noising. We exploit the recently proposed tensor-Singular Value Decomposition (t-SVD)[11]. Based on t-SVD, the notion of multilinear rank and a related tensor nuclear norm was proposed in [11] to characterize informational and structural complexity of multilinear data. We first show that videos with linear camera motion can be represented more efficiently using t-SVD compared to the approaches based on vectorizing or flattening of the tensors. Since efficiency in representation implies efficiency in recovery, we outline a tensor nuclear norm penalized algorithm for video completion from missing entries. Application of the proposed algorithm for video recovery from missing entries is shown to yield a superior performance over existing methods. We also consider the problem of tensor robust Principal Component Analysis (PCA) for de-noising 3-D video data from sparse random corruptions. We show superior performance of our method compared to the matrix robust PCA adapted to this setting as proposed in [4].

8 pages, 8 figures. It is accepted as CVPR 2014 oral presentation. arXiv admin note: substantial text overlap with arXiv:1307.0805

References in corpus (1)

• Near-Optimal Algorithms for Online Matrix Prediction

Cited by in corpus (16)

• Dynamic Graph Convolutional Networks Using the Tensor M-Product
• Tensor Robust Principal Component Analysis: Exact Recovery of Corrupted Low-Rank Tensors via Convex Optimization
• Fourth-order Tensors with Multidimensional Discrete Transforms
• Tensor Completion Algorithms in Big Data Analytics
• Exact Tensor Completion from Sparsely Corrupted Observations via Convex Optimization
• Fast and Accurate Tensor Completion with Total Variation Regularized Tensor Trains
• A Fast Algorithm for Cosine Transform Based Tensor Singular Value Decomposition
• General Data Analytics with Applications to Visual Information Analysis: A Provable Backward-Compatible Semisimple Paradigm over T-Algebra
• Effective Image Retrieval via Multilinear Multi-index Fusion
• A Randomized Tensor Singular Value Decomposition based on the t-product
• Low-Rank Tensor Completion by Truncated Nuclear Norm Regularization
• MARLow: A Joint Multiplanar Autoregressive and Low-Rank Approach for Image Completion
• A Tensor Completion Approach for Efficient and Robust Fingerprint-based Indoor Localization
• Self-supervised Knowledge Distillation Using Singular Value Decomposition
• Sturm: Sparse Tubal-Regularized Multilinear Regression for fMRI
• Augmented Robust PCA For Foreground-Background Separation on Noisy, Moving Camera Video