• Corpus ID: 247762757

# Coarse to Fine Two-Stage Approach to Robust Tensor Completion of Visual Data

@inproceedings{He2021CoarseTF,
title={Coarse to Fine Two-Stage Approach to Robust Tensor Completion of Visual Data},
author={Yicong He and George K. Atia},
year={2021}
}
• Published 19 June 2021
• Computer Science
—Tensor completion is the problem of estimating the missing values of high-order data from partially observed entries. Data corruption due to prevailing outliers poses major challenges to traditional tensor completion algorithms, which catalyzed the development of robust tensor completion algorithms that alleviate the effect of outliers. However, existing robust methods largely presume that the corruption is sparse, which may not hold in practice. In this paper, we develop a two-stage robust…

## References

SHOWING 1-10 OF 44 REFERENCES
Tensor-ring Nuclear Norm Minimization and Application for Visual : Data Completion
• Computer Science
ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
• 2019
This work introduces a class of new tensor nuclear norms by using tensor circular unfolding and theoretically establishes connection between the rank of the circularly unfolded matrices and the TR ranks, and develops an efficient tensor completion algorithm by minimizing the proposed Tensor nuclear norm.
Robust Low-Rank Tensor Recovery With Regularized Redescending M-Estimator
• Computer Science, Mathematics
IEEE Transactions on Neural Networks and Learning Systems
• 2016
This paper proposes approaches based on the regularized redescending M-estimators, which have been introduced in robust statistics, and develops algorithms based on proximal and linearized block coordinate descent methods that are robust to outliers and effective in the absence of outliers.
Low Tensor-Ring Rank Completion by Parallel Matrix Factorization
• Computer Science
IEEE Transactions on Neural Networks and Learning Systems
• 2021
This article introduces a balanced unfolding operation called tensor circular unfolding, by which the relationship between TR rank and the ranks of tensor unfoldings is theoretically established, and proposes an algorithm to exploit the low TR-rank structure by performing parallel low-rank matrix factorizations to all circularly unfolded matrices.
Nonlocal Low-Rank Tensor Completion for Visual Data
• Computer Science
IEEE Transactions on Cybernetics
• 2021
The theoretical analysis shows that the error caused by Patch Mismatch can be decomposed into two components, one of which can be bounded by a reasonable assumption named local patch similarity, and the other part is lower than that using matrix completion.
Robust Low-Rank Tensor Completion Based on Tensor Ring Rank via $\ell _{p,\epsilon }$-Norm
• Computer Science
IEEE Transactions on Signal Processing
• 2021
Experimental results from applications of image inpainting, video restoration and target estimation demonstrate that the robust tensor completion approach outperforms state-of-the-art methods in terms of recovery accuracy.
Tensor train rank minimization with nonlocal self-similarity for tensor completion
• Computer Science
Inverse Problems & Imaging
• 2021
The alternating direction method of multipliers tailored for the specific structure to solve the proposed model is developed and is superior to several existing state-of-the-art methods in terms of both qualitative and quantitative measures.
Hierarchical Tensor Ring Completion
• Computer Science
ArXiv
• 2020
A hierarchical tensor ring decomposition for more compact representation that outperforms state-of-the-arts ones in terms of recovery accuracy and total variation to exploit the locally piece-wise smoothness data structure is proposed.
Provable tensor ring completion
• Signal Processing, vol. 171, p. 107486, 2020.
• 2020