Deep Learning for Image Super-Resolution: A Survey (2019-02-16T00:00:00.000000Z)

TL;DR

A survey on recent advances of image super-resolution techniques using deep learning approaches in a systematic way, which can roughly group the existing studies of SR techniques into three major categories: supervised SR, unsupervised SR, and domain-specific SR.

Abstract

Image Super-Resolution (SR) is an important class of image processing techniqueso enhance the resolution of images and videos in computer vision. Recent years have witnessed remarkable progress of image super-resolution using deep learning techniques. This article aims to provide a comprehensive survey on recent advances of image super-resolution using deep learning approaches. In general, we can roughly group the existing studies of SR techniques into three major categories: supervised SR, unsupervised SR, and domain-specific SR. In addition, we also cover some other important issues, such as publicly available benchmark datasets and performance evaluation metrics. Finally, we conclude this survey by highlighting several future directions and open issues which should be further addressed by the community in the future.

Authors

S. Hoi

19 papers

Zhihao Wang

1 papers

Jian Chen

1 papers

TL;DR

Abstract

Authors

References234 items

Pixel Transposed Convolutional Networks

China

NAT: Neural Architecture Transformer for Accurate and Compact Architectures

3D Appearance Super-Resolution With Deep Learning

Hyperspectral Image Super-Resolution With Optimized RGB Guidance

Second-Order Attention Network for Single Image Super-Resolution

Residual Networks for Light Field Image Super-Resolution

Towards Real Scene Super-Resolution With Raw Images

Zoom to Learn, Learn to Zoom

Camera Lens Super-Resolution

Feedback Network for Image Super-Resolution

Learning Parallax Attention for Stereo Image Super-Resolution

Meta-SR: A Magnification-Arbitrary Network for Super-Resolution

Residual Non-local Attention Networks for Image Restoration

Deep Network Interpolation for Continuous Imagery Effect Transition

2018 PIRM Challenge on Perceptual Image Super-resolution

Dual Reconstruction Nets for Image Super-Resolution with Gradient Sensitive Loss

Deep Learning-based Image Super-Resolution Considering Quantitative and Perceptual Quality

PIRM Challenge on Perceptual Image Enhancement on Smartphones: Report

Super-Identity Convolutional Neural Network for Face Hallucination

SRFeat: Single Image Super-Resolution with Feature Discrimination

Analyzing Perception-Distortion Tradeoff using Enhanced Perceptual Super-resolution Network

Multi-scale Residual Network for Image Super-Resolution

Face Super-Resolution Guided by Facial Component Heatmaps

Generative adversarial network-based image super-resolution using perceptual content losses

SOD-MTGAN: Small Object Detection via Multi-Task Generative Adversarial Network

Fast and Efficient Image Quality Enhancement via Desubpixel Convolutional Neural Networks

The 2018 PIRM Challenge on Perceptual Image Super-Resolution

ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks

To learn image super-resolution, use a GAN to learn how to do image degradation first

Image Super-Resolution Using Very Deep Residual Channel Attention Networks

The relativistic discriminator: a key element missing from standard GAN

DARTS: Differentiable Architecture Search

Unsupervised Image Super-Resolution Using Cycle-in-Cycle Generative Adversarial Networks

Enhancing the Spatial Resolution of Stereo Images Using a Parallax Prior

Fight Ill-Posedness with Ill-Posedness: Single-shot Variational Depth Super-Resolution from Shading

NTIRE 2018 Challenge on Image Dehazing: Methods and Results

Super-Resolving Very Low-Resolution Face Images with Supplementary Attributes

Deep Video Super-Resolution Network Using Dynamic Upsampling Filters Without Explicit Motion Compensation

Efficient Module Based Single Image Super Resolution for Multiple Problems

Persistent Memory Residual Network for Single Image Super Resolution

Attribute Augmented Convolutional Neural Network for Face Hallucination

Image Super-Resolution via Progressive Cascading Residual Network

Feature Super-Resolution: Make Machine See More Clearly

Hallucinating Compressed Face Images

Multi-level Wavelet-CNN for Image Restoration

Convolutional Neural Network-Based Video Super-Resolution for Action Recognition

New Techniques for Preserving Global Structure and Denoising with Low Information Loss in Single-Image Super-Resolution

Image Super-Resolution via Dual-State Recurrent Networks

Unsupervised Sparse Dirichlet-Net for Hyperspectral Image Super-Resolution

A Fully Progressive Approach to Single-Image Super-Resolution

Recovering Realistic Texture in Image Super-Resolution by Deep Spatial Feature Transform

Video Super-Resolution via Bidirectional Recurrent Convolutional Networks

Task-Driven Super Resolution: Object Detection in Low-resolution Images

Learning Temporal Dynamics for Video Super-Resolution: A Deep Learning Approach

Fast and Accurate Single Image Super-Resolution via Information Distillation Network

Fast, Accurate, and, Lightweight Super-Resolution with Cascading Residual Network

Depth-aware CNN for RGB-D Segmentation

Deep Back-Projection Networks for Super-Resolution

Residual Dense Network for Image Super-Resolution

Spectral Normalization for Generative Adversarial Networks

Efficient Neural Architecture Search via Parameter Sharing

Frame-Recurrent Video Super-Resolution

The Unreasonable Effectiveness of Deep Features as a Perceptual Metric

Joint convolutional neural pyramid for depth map super-resolution

Learning a Single Convolutional Super-Resolution Network for Multiple Degradations

Zero-Shot Super-Resolution Using Deep Internal Learning

Super-FAN: Integrated Facial Landmark Localization and Super-Resolution of Real-World Low Resolution Faces in Arbitrary Poses with GANs

Deep Image Prior

FSRNet: End-to-End Learning Face Super-Resolution with Facial Priors

V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation from a Single Depth Map

xUnit: Learning a Spatial Activation Function for Efficient Image Restoration

The Perception-Distortion Tradeoff

Single-Image Super-Resolution for Remote Sensing Data Using Deep Residual-Learning Neural Network

Progressive Growing of GANs for Improved Quality, Stability, and Variation

Fast and Accurate Image Super-Resolution with Deep Laplacian Pyramid Networks