Deep Networks for Image Super-Resolution with Sparse Prior

Abstract

Deep learning techniques have been successfully applied in many areas of computer vision, including low-level image restoration problems. For image super-resolution, several models based on deep neural networks have been recently proposed and attained superior performance that overshadows all previous handcrafted models. The question then arises whether large-capacity and data-driven models have become the dominant solution to the ill-posed super-resolution problem. In this paper, we argue that domain expertise represented by the conventional sparse coding model is still valuable, and it can be combined with the key ingredients of deep learning to achieve further improved results. We show that a sparse coding model particularly designed for super-resolution can be incarnated as a neural network, and trained in a cascaded structure from end to end. The interpretation of the network based on sparse coding leads to much more efficient and effective training, as well as a reduced model size. Our model is evaluated on a wide range of images, and shows clear advantage over existing state-of-the-art methods in terms of both restoration accuracy and human subjective quality.

References in corpus (1)

• Fast Inference in Sparse Coding Algorithms with Applications to Object Recognition

Cited by in corpus (33)

• The Unreasonable Effectiveness of Deep Features as a Perceptual Metric
• Deep-STORM: super-resolution single-molecule microscopy by deep learning
• DeepNIS: Deep Neural Network for Nonlinear Electromagnetic Inverse Scattering
• Non-Local Recurrent Network for Image Restoration
• Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network
• MemNet: A Persistent Memory Network for Image Restoration
• Learned SPARCOM: Unfolded Deep Super-Resolution Microscopy
• Fast and Accurate Image Super-Resolution with Deep Laplacian Pyramid Networks
• End-to-End Image Super-Resolution via Deep and Shallow Convolutional Networks
• Face Recognition in Low Quality Images: A Survey
• Studying Very Low Resolution Recognition Using Deep Networks
• Image Super-Resolution via Dual-State Recurrent Networks
• Deep Embedding Convolutional Neural Network for Synthesizing CT Image from T1-Weighted MR Image
• Dual Reconstruction Nets for Image Super-Resolution with Gradient Sensitive Loss
• Single Image Super-resolution via a Lightweight Residual Convolutional Neural Network
• Single Image Super-Resolution with Dilated Convolution based Multi-Scale Information Learning Inception Module
• Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution
• End-to-End Kernel Learning with Supervised Convolutional Kernel Networks
• Learning Fully Convolutional Networks for Iterative Non-blind Deconvolution
• EPSR: Edge Profile Super resolution
• ${\mathbf{D}}^3$: Deep Dual-Domain Based Fast Restoration of JPEG-Compressed Images
• Texture Hallucination for Large-Factor Painting Super-Resolution
• Local- and Holistic- Structure Preserving Image Super Resolution via Deep Joint Component Learning
• DeepDPM: Dynamic Population Mapping via Deep Neural Network
• Deep Inception-Residual Laplacian Pyramid Networks for Accurate Single Image Super-Resolution
• Generic 3D Convolutional Fusion for image restoration
• CRNet: Image Super-Resolution Using A Convolutional Sparse Coding Inspired Network
• Can Deep Learning Relax Endomicroscopy Hardware Miniaturization Requirements?
• Tensor-Generative Adversarial Network with Two-dimensional Sparse Coding: Application to Real-time Indoor Localization
• Content-adaptive Representation Learning for Fast Image Super-resolution
• Optical Flow Super-Resolution Based on Image Guidence Using Convolutional Neural Network
• Joint Face Hallucination and Deblurring via Structure Generation and Detail Enhancement
• Human-Aware Motion Deblurring