Deep Plug-and-play Prior for Low-rank Tensor Completion

Abstract

Multi-dimensional images, such as color images and multi-spectral images, are highly correlated and contain abundant spatial and spectral information. However, real-world multi-dimensional images are usually corrupted by missing entries. By integrating deterministic low-rankness prior to the data-driven deep prior, we suggest a novel regularized tensor completion model for multi-dimensional image completion. In the objective function, we adopt the newly emerged tensor nuclear norm (TNN) to characterize the global low-rankness prior of the multi-dimensional images. We also formulate an implicit regularizer by plugging into a denoising neural network (termed as deep denoiser), which is convinced to express the deep image prior learned from a large number of natural images. The resulting model can be solved by the alternating directional method of multipliers algorithm under the plug-and-play (PnP) framework. Experimental results on color images, videos, and multi-spectral images demonstrate that the proposed method can recover both the global structure and fine details very well and achieve superior performance over competing methods in terms of quality metrics and visual effects.

Accepted by Neurocoputing. Code availiable at https://github.com/TaiXiangJiang/Deep-Plug-and-Play-Prior-for-Low-Rank-Tensor-Completion

References in corpus (8)

• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• FFDNet: Toward a Fast and Flexible Solution for CNN based Image Denoising
• Tensor Decomposition for Signal Processing and Machine Learning
• Bayesian CP Factorization of Incomplete Tensors with Automatic Rank Determination
• Fast Hyperspectral Image Denoising and Inpainting Based on Low-Rank and Sparse Representations
• Plug-and-Play Priors for Bright Field Electron Tomography and Sparse Interpolation
• Multi-dimensional imaging data recovery via minimizing the partial sum of tubal nuclear norm
• Plug-and-Play Methods Provably Converge with Properly Trained Denoisers

Cited by in corpus (5)

• Tensor Recovery Based on A Novel Non-convex Function Minimax Logarithmic Concave Penalty Function
• Dictionary Learning with Low-rank Coding Coefficients for Tensor Completion
• Salt and pepper noise removal method based on stationary Framelet transform with non-convex sparsity regularization
• Tensor Full Feature Measure and Its Nonconvex Relaxation Applications to Tensor Recovery
• Tensor Completion via Convolutional Sparse Coding Regularization