GaintPandaSR

A Image Super-Resolution Codebase which implements all sota approaches.

1. Dataset

Public Datasets:
   |- DIV2K
   |- Flicker2K
   |- DIV8K
   |- Set5

2. Code Role

We use the PEP8 style and we should add docstring to every new function.

3. Folder structure

. ├── data │   └── init.py ├── log │   └── init.py ├── metrics │   └── init.py ├── model │   └── init.py ├── README.md ├── requirements.txt ├── tools │   ├── demo.py │   ├── test.py │   └── train.py └── utils └── init.py

4. Papers & Codes

Name	Summary	Paper	Code
2015
SRCNN	Image Super-Resolution Using Deep Convolutional Networks	[arXiv]	[code]
2016
SRGAN	Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network	[arXiv]	[code]
FSRGAN	Accelerating the Super-Resolution Convolutional Neural Network	[arXiv]	[code]
EnhanceNet	Single Image Super-Resolution Through Automated Texture Synthesis	[arXiv]	[code]
2017
LapSRN	Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution	[arXiv]	[code]
EDSR	Enhanced Deep Residual Networks for Single Image Super-Resolution	[arXiv]	[code]
2018