Learning Deep CNN Denoiser Prior for Image Restoration

Abstract

Model-based optimization methods and discriminative learning methods have been the two dominant strategies for solving various inverse problems in low-level vision. Typically, those two kinds of methods have their respective merits and drawbacks, e.g., model-based optimization methods are flexible for handling different inverse problems but are usually time-consuming with sophisticated priors for the purpose of good performance; in the meanwhile, discriminative learning methods have fast testing speed but their application range is greatly restricted by the specialized task. Recent works have revealed that, with the aid of variable splitting techniques, denoiser prior can be plugged in as a modular part of model-based optimization methods to solve other inverse problems (e.g., deblurring). Such an integration induces considerable advantage when the denoiser is obtained via discriminative learning. However, the study of integration with fast discriminative denoiser prior is still lacking. To this end, this paper aims to train a set of fast and effective CNN (convolutional neural network) denoisers and integrate them into model-based optimization method to solve other inverse problems. Experimental results demonstrate that the learned set of denoisers not only achieve promising Gaussian denoising results but also can be used as prior to deliver good performance for various low-level vision applications.

Accepted to CVPR 2017. Code: https://github.com/cszn/ircnn

Cited by in corpus (44)

• Image Super-Resolution Using Very Deep Residual Channel Attention Networks
• DeepISP: Towards Learning an End-to-End Image Processing Pipeline
• Non-Local Recurrent Network for Image Restoration
• Residual Dense Network for Image Super-Resolution
• An Online Plug-and-Play Algorithm for Regularized Image Reconstruction
• RARE: Image Reconstruction using Deep Priors Learned without Ground Truth
• Deep Chain HDRI: Reconstructing a High Dynamic Range Image from a Single Low Dynamic Range Image
• Spatial-Spectral Fusion by Combining Deep Learning and Variation Model
• Deep Mean-Shift Priors for Image Restoration
• Learning a Single Convolutional Super-Resolution Network for Multiple Degradations
• Unified Single-Image and Video Super-Resolution via Denoising Algorithms
• Three-dimensional Optical Coherence Tomography Image Denoising through Multi-input Fully-Convolutional Networks
• Multi-level Wavelet-CNN for Image Restoration
• Hallucinated-IQA: No-Reference Image Quality Assessment via Adversarial Learning
• Convolutional Recurrent Neural Networks for Dynamic MR Image Reconstruction
• Learning a Discriminative Prior for Blind Image Deblurring
• A Single Video Super-Resolution GAN for Multiple Downsampling Operators based on Pseudo-Inverse Image Formation Models
• Multi-feature super-resolution network for cloth wrinkle synthesis
• When AWGN-based Denoiser Meets Real Noises
• Data-driven geophysics: from dictionary learning to deep learning
• Image Obfuscation for Privacy-Preserving Machine Learning
• CRRN: Multi-Scale Guided Concurrent Reflection Removal Network
• Deep Learning Techniques for Inverse Problems in Imaging
• Multi-bin Trainable Linear Unit for Fast Image Restoration Networks
• Deep Blind Image Inpainting
• Weighted Sigmoid Gate Unit for an Activation Function of Deep Neural Network
• Image Correction via Deep Reciprocating HDR Transformation
• Deep Learning for Low-Dose CT Denoising
• Deep Network Interpolation for Continuous Imagery Effect Transition
• LSD${}_2$ -- Joint Denoising and Deblurring of Short and Long Exposure Images with CNNs
• Image Demoireing with Learnable Bandpass Filters
• Pyramidal Dense Attention Networks for Lightweight Image Super-Resolution
• Dense xUnit Networks
• Classifying degraded images over various levels of degradation
• KRNET: Image Denoising with Kernel Regulation Network
• A Group Variational Transformation Neural Network for Fractional Interpolation of Video Coding
• Deep Gradient Projection Networks for Pan-sharpening
• RBSRICNN: Raw Burst Super-Resolution through Iterative Convolutional Neural Network
• Extra Proximal-Gradient Inspired Non-local Network
• A Deep Optimization Approach for Image Deconvolution
• Iterative Residual Image Deconvolution
• Learning to Learn Graph Topologies
• Learning Deep Image Priors for Blind Image Denoising
• Learning Collaborative Generation Correction Modules for Blind Image Deblurring and Beyond