Nonlinear transform induced tensor nuclear norm for tensor completion
From MaRDI portal
Publication:2162239
Abstract: The linear transform-based tensor nuclear norm (TNN) methods have recently obtained promising results for tensor completion. The main idea of this type of methods is exploiting the low-rank structure of frontal slices of the targeted tensor under the linear transform along the third mode. However, the low-rankness of frontal slices is not significant under linear transforms family. To better pursue the low-rank approximation, we propose a nonlinear transform-based TNN (NTTNN). More concretely, the proposed nonlinear transform is a composite transform consisting of the linear semi-orthogonal transform along the third mode and the element-wise nonlinear transform on frontal slices of the tensor under the linear semi-orthogonal transform, which are indispensable and complementary in the composite transform to fully exploit the underlying low-rankness. Based on the suggested low-rankness metric, i.e., NTTNN, we propose a low-rank tensor completion (LRTC) model. To tackle the resulting nonlinear and nonconvex optimization model, we elaborately design the proximal alternating minimization (PAM) algorithm and establish the theoretical convergence guarantee of the PAM algorithm. Extensive experimental results on hyperspectral images, multispectral images, and videos show that the our method outperforms linear transform-based state-of-the-art LRTC methods qualitatively and quantitatively.
Recommendations
- A Learnable Group-Tube Transform Induced Tensor Nuclear Norm and Its Application for Tensor Completion
- Tensor train rank minimization with nonlocal self-similarity for tensor completion
- A non-convex tensor rank approximation for tensor completion
- Tensor completion via a generalized transformed tensor t-product decomposition without t-SVD
- Tensor completion based on triple tubal nuclear norm
Cites work
- A non-convex tensor rank approximation for tensor completion
- A preconditioner based on a splitting-type iteration method for solving complex symmetric indefinite linear systems
- A Riemannian optimization approach for solving the generalized eigenvalue problem for nonsquare matrix pencils
- A Singular Value Thresholding Algorithm for Matrix Completion
- An ^2-^q regularization method for large discrete ill-posed problems
- An efficient randomized algorithm for computing the approximate Tucker decomposition
- Convergence of descent methods for semi-algebraic and tame problems: proximal algorithms, forward-backward splitting, and regularized Gauss-Seidel methods
- Efficient Tensor Completion for Color Image and Video Recovery: Low-Rank Tensor Train
- Factorization strategies for third-order tensors
- Framelet Representation of Tensor Nuclear Norm for Third-Order Tensor Completion
- Global convergence of ADMM in nonconvex nonsmooth optimization
- scientific article; zbMATH DE number 51537 (Why is no real title available?)
- Low-rank tensor completion based on log-det rank approximation and matrix factorization
- Low-rank tensor completion using matrix factorization based on tensor train rank and total variation
- Most tensor problems are NP-hard
- Nonconvex optimization for robust tensor completion from grossly sparse observations
- Proximal alternating linearized minimization for nonconvex and nonsmooth problems
- Proximal Alternating Minimization and Projection Methods for Nonconvex Problems: An Approach Based on the Kurdyka-Łojasiewicz Inequality
- Robust tensor completion using transformed tensor singular value decomposition
- Tensor completion via fully-connected tensor network decomposition with regularized factors
- Tensor Decompositions and Applications
- Tensor Q-rank: new data dependent definition of tensor rank
- Tensor-tensor products with invertible linear transforms
- Tensor-train decomposition
- The relaxation convergence of multisplitting AOR method for linear complementarity problem
- The singular value decomposition: anatomy of optimizing an algorithm for extreme scale
- Third-order tensors as operators on matrices: a theoretical and computational framework with applications in imaging
Cited in
(22)- Tensor train rank minimization with nonlocal self-similarity for tensor completion
- Reshaped tensor nuclear norms for higher order tensor completion
- Tensor completion via a generalized transformed tensor t-product decomposition without t-SVD
- Multi-dimensional image recovery via fully-connected tensor network decomposition under the learnable transforms
- Tensor completion based on triple tubal nuclear norm
- Hankel Matrix Nuclear Norm Regularized Tensor Completion for N-dimensional Exponential Signals
- Fully-connected tensor network decomposition for robust tensor completion problem
- Tensor factorization via transformed tensor-tensor product for image alignment
- A Learnable Group-Tube Transform Induced Tensor Nuclear Norm and Its Application for Tensor Completion
- Nonconvex optimization for third‐order tensor completion under wavelet transform
- Multi-dimensional image recovery via self-supervised nonlinear transform based a three-directional tensor nuclear norm
- Grouped tensor low-rank representation for multi-dimensional image recovery
- Importance-aware nonlocal tensor nuclear norm for high-dimensional image recovery
- Dictionary-based block term decomposition for third-order tensors
- Inconsistent tensor-tensor product for low-rank tensor factorization
- A novel tensor decomposition-based approach for internet traffic data recovery and forecast
- Projected tensor-tensor products for efficient computation of optimal multiway data representations
- Slice-wise irregular tensor decomposition and its application
- Efficient quaternion CUR method for low-rank approximation to quaternion matrix
- Tensor robust principal component analysis with total generalized variation for high-dimensional data recovery
- Doubly robust logistic regression for image classification
- Spectral-spatial extraction through layered tensor decomposition for hyperspectral anomaly detection
This page was built for publication: Nonlinear transform induced tensor nuclear norm for tensor completion
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2162239)
