Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Systematic Survey of Deep Learning-based Single-Image Super-Resolution

Juncheng Li, Zehua Pei, Wenjie Li, Guangwei Gao, Longguang Wang, Yingqian Wang, Tieyong Zeng

TL;DR

This survey delineates the DL-based single-image super-resolution (SISR) landscape by organizing methods into Simulation SISR, Real-World SISR, and Domain-Specific Applications. It details problem formulation, degradation models, upsampling strategies, loss functions, and evaluation metrics, and then surveys networks designed for efficiency, perceptual quality, and information utilization. The authors compare reconstruction results across 53 methods, highlighting the prominence of large datasets and Transformer-based models while noting the ongoing need for lightweight, scale-arbitrary, and real-world robust solutions. They also identify remaining challenges—edge-device feasibility, flexible architectures, novel losses, and integration with high-level vision tasks—and propose directions for real-world impact and future research. Overall, the work provides a comprehensive, target-oriented map of DL-based SISR and sets the stage for continued advances toward practical, high-quality SR in diverse applications.

Abstract

Single-image super-resolution (SISR) is an important task in image processing, which aims to enhance the resolution of imaging systems. Recently, SISR has made a huge leap and has achieved promising results with the help of deep learning (DL). In this survey, we give an overview of DL-based SISR methods and group them according to their design targets. Specifically, we first introduce the problem definition, research background, and the significance of SISR. Secondly, we introduce some related works, including benchmark datasets, upsampling methods, optimization objectives, and image quality assessment methods. Thirdly, we provide a detailed investigation of SISR and give some domain-specific applications of it. Fourthly, we present the reconstruction results of some classic SISR methods to intuitively know their performance. Finally, we discuss some issues that still exist in SISR and summarize some new trends and future directions. This is an exhaustive survey of SISR, which can help researchers better understand SISR and inspire more exciting research in this field. An investigation project for SISR is provided at https://github.com/CV-JunchengLi/SISR-Survey.

A Systematic Survey of Deep Learning-based Single-Image Super-Resolution

TL;DR

This survey delineates the DL-based single-image super-resolution (SISR) landscape by organizing methods into Simulation SISR, Real-World SISR, and Domain-Specific Applications. It details problem formulation, degradation models, upsampling strategies, loss functions, and evaluation metrics, and then surveys networks designed for efficiency, perceptual quality, and information utilization. The authors compare reconstruction results across 53 methods, highlighting the prominence of large datasets and Transformer-based models while noting the ongoing need for lightweight, scale-arbitrary, and real-world robust solutions. They also identify remaining challenges—edge-device feasibility, flexible architectures, novel losses, and integration with high-level vision tasks—and propose directions for real-world impact and future research. Overall, the work provides a comprehensive, target-oriented map of DL-based SISR and sets the stage for continued advances toward practical, high-quality SR in diverse applications.

Abstract

Single-image super-resolution (SISR) is an important task in image processing, which aims to enhance the resolution of imaging systems. Recently, SISR has made a huge leap and has achieved promising results with the help of deep learning (DL). In this survey, we give an overview of DL-based SISR methods and group them according to their design targets. Specifically, we first introduce the problem definition, research background, and the significance of SISR. Secondly, we introduce some related works, including benchmark datasets, upsampling methods, optimization objectives, and image quality assessment methods. Thirdly, we provide a detailed investigation of SISR and give some domain-specific applications of it. Fourthly, we present the reconstruction results of some classic SISR methods to intuitively know their performance. Finally, we discuss some issues that still exist in SISR and summarize some new trends and future directions. This is an exhaustive survey of SISR, which can help researchers better understand SISR and inspire more exciting research in this field. An investigation project for SISR is provided at https://github.com/CV-JunchengLi/SISR-Survey.

Paper Structure

This paper contains 42 sections, 23 equations, 12 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Problem Setting and Related Works
  3. Problem Definition
  4. Benchmark Datasets
  5. Degradation Mode
  6. Training and Test Datasets
  7. Upsampling Methods
  8. Interpolation Methods
  9. Transposed Convolutional Layers
  10. Sub-pixel Convolutional Layer
  11. Optimization Objective
  12. Learning Strategy
  13. Loss Function
  14. Assessment Methods
  15. Image Reconstruction Accuracy
  16. ...and 27 more sections

Figures (12)

  • Figure 1: SISR aims to reconstruct a high-resolution (HR) image from its degraded low-resolution (LR) one.
  • Figure 2: The content and taxonomy of this survey. In this survey, we divide image super-resolution methods into three categories: Simulation SISR, Real-World SISR, and Domain-Specific Applications.
  • Figure 3: The training process of data-driven based deep neural networks.
  • Figure 4: Upsampling methods: (a) transposed convolutional layers (b) sub-pixel convolutional layer.
  • Figure 5: Sketch of residual learning architecture / residual block.
  • ...and 7 more figures