Abstract:Most existing methods usually formulate the non-blind deconvolution problem into a maximum-a-posteriori framework and address it by manually designing a variety of regularization terms and data terms of the latent clear images. However, explicitly designing these two terms is quite challenging and usually leads to complex optimization problems which are difficult to solve. This paper proposes an effective non-blind deconvolution approach by learning discriminative shrinkage functions to model these terms implicitly. Most existing methods use deep convolutional neural networks (CNNs) or radial basis functions to learn the regularization term simply. In contrast, we formulate both the data term and regularization term and split the deconvolution model into data-related and regularization-related sub-problems according to the alternating direction method of multipliers. We explore the properties of the Maxout function and develop a deep CNN model with Maxout layers to learn discriminative shrinkage functions, which directly approximates the solutions of these two sub-problems. Moreover, the fast-Fourier-transform-based image restoration usually leads to ringing artifacts. At the same time, the conjugate-gradient-based approach is time-consuming; we develop the Conjugate Gradient Network to restore the latent clear images effectively and efficiently. Experimental results show that the proposed method performs favorably against the state-of-the-art methods in terms of efficiency and accuracy. Source codes, models, and more results are available at https://github.com/setsunil/DSDNet.

A Deconvolution Methods Based on Retinex Prior.

Multi-frame Blind Deconvolution Using Sparse Priors

Non-blind image deconvolution using natural image gradient prior

High quality non-blind image deconvolution using the Fields of Experts prior

Blind Deconvolution Using an Improved L0 Sparse Representation

An Improved Image Deconvolution Approach Using Local Constraint

A Novel Framework Method for Non-Blind Deconvolution Using Subspace Images Priors

High-quality Blind Defocus Deblurring of Multispectral Images with Optics and Gradient Prior.

A Deep Optimization Approach for Image Deconvolution.

High Quality Image-Pair-based Deblurring Method Using Edge Mask and Improved Residual Deconvolution

Blind multi-Poissonian image deconvolution with sparse log-step gradient prior

Iteratively Reweighted Blind Deconvolution With Adaptive Regularization Parameter Estimation.

A Convex Optimization Model and Algorithm for Retinex

Learning Discriminative Shrinkage Deep Networks for Image Deconvolution

Blind Image Deblurring Using Adaptive Priors.

Blind Image Deconvolution Assisted By Robust Non-Gaussian Priors

Tuning-free Plug-and-Play Hyperspectral Image Deconvolution with Deep Priors

Subspace-based non-blind deconvolution.

Blind Image Deconvolution Using Variational Deep Image Prior

An improved dehazing algorithm based on dark channel prior

Multi-Parseval Frame–based Nonconvex Sparse Image Deconvolution