基于深度学习的图像超分辨率文献和常用数据

小鱼儿 2022-10-29 09:14 182阅读 0赞

### 文章目录 ###

*  文献
 *  常用数据

未完成

# 文献 #

<table> 
 <thead> 
  <tr> 
   <th>论文简称</th> 
   <th>论文名称</th> 
   <th>文章地址</th> 
   <th>源码地址</th> 
  </tr> 
 </thead> 
 <tbody> 
  <tr> 
   <td>SRCNN</td> 
   <td>Image Super-Resolution Using Deep Convolutional Networks</td> 
   <td>https://arxiv.org/abs/1501.00092</td> 
   <td>http://mmlab.ie.cuhk.edu.hk/projects/SRCNN/SRCNN_train.zip (caffe)，<br>http://mmlab.ie.cuhk.edu.hk/projects/SRCNN/SRCNN_v1.zip (MATLAB),<br>https://github.com/tegg89/SRCNN-Tensorflow (TensorFlow，非作者)</td> 
  </tr> 
  <tr> 
   <td>FSRCNN</td> 
   <td>Accelerating the Super-Resolution Convolutional Neural Network</td> 
   <td>https://arxiv.org/abs/1608.00367</td> 
   <td>https://drive.google.com/open?id=0B7tU5Pj1dfCMWjhhaE1HR3dqcGs (caffe), <br>https://drive.google.com/open?id=0B7tU5Pj1dfCMVktYZUN2aV8xVTQ (MATLAB)</td> 
  </tr> 
  <tr> 
   <td>VDSR</td> 
   <td>Accurate Image Super-Resolution Using Very Deep Convolutional Networks</td> 
   <td>https://arxiv.org/abs/1511.04587</td> 
   <td>https://github.com/twtygqyy/pytorch-vdsr（PyTorch，非作者）</td> 
  </tr> 
  <tr> 
   <td>ESPCN</td> 
   <td>Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network</td> 
   <td>https://arxiv.org/abs/1609.05158</td> 
   <td>https://github.com/leftthomas/ESPCN（PyTorch，非作者）</td> 
  </tr> 
  <tr> 
   <td>LapSRN</td> 
   <td>Fast and Accurate Image Super-Resolution with Deep Laplacian Pyramid Networks</td> 
   <td>https://arxiv.org/abs/1710.01992</td> 
   <td>https://github.com/phoenix104104/LapSRN (MATLAB), <br>https://github.com/twtygqyy/pytorch-LapSRN （PyTorch，非作者）</td> 
  </tr> 
  <tr> 
   <td>EDSR</td> 
   <td>Enhanced Deep Residual Networks for Single Image Super-Resolution</td> 
   <td>https://arxiv.org/abs/1707.02921</td> 
   <td>https://github.com/LimBee/NTIRE2017 (torch), <br>https://github.com/thstkdgus35/EDSR-PyTorch (PyTorch)</td> 
  </tr> 
  <tr> 
   <td>RCAN</td> 
   <td>Image Super-Resolution Using Very Deep Residual Channel Attention Networks</td> 
   <td>https://arxiv.org/abs/1807.02758</td> 
   <td>https://github.com/yulunzhang/RCAN (PyTorch)</td> 
  </tr> 
  <tr> 
   <td>SRGAN</td> 
   <td>Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network</td> 
   <td>https://arxiv.org/abs/1609.04802</td> 
   <td>https://github.com/tensorlayer/srgan （TensorFlow，非作者），<br>https://github.com/leftthomas/SRGAN （PyTorch，非作者）</td> 
  </tr> 
  <tr> 
   <td>ESRGAN</td> 
   <td>ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks</td> 
   <td>https://arxiv.org/abs/1809.00219</td> 
   <td>https://github.com/xinntao/ESRGAN (PyTorch), <br>https://github.com/xinntao/BasicSR (PyTorch)</td> 
  </tr> 
  <tr> 
   <td>SFTGAN</td> 
   <td>Recovering Realistic Texture in Image Super-resolution by Deep Spatial Feature Transform</td> 
   <td>https://arxiv.org/abs/1804.02815</td> 
   <td>https://github.com/xinntao/SFTGAN (PyTorch), <br>https://github.com/xinntao/BasicSR (PyTorch)</td> 
  </tr> 
  <tr> 
   <td>Meta-SR</td> 
   <td>Meta-SR: A Magnification-Arbitrary Network for Super-Resolution</td> 
   <td>https://arxiv.org/abs/1903.00875</td> 
   <td>https://github.com/XuecaiHu/Meta-SR-Pytorch</td> 
  </tr> 
  <tr> 
   <td>DRN</td> 
   <td>Closed-loop Matters: Dual Regression Networks for Single Image Super-Resolution</td> 
   <td>https://arxiv.org/abs/2003.07018</td> 
   <td>https://github.com/guoyongcs/DRN</td> 
  </tr> 
  <tr> 
   <td>CutBlur</td> 
   <td>Rethinking Data Augmentation for Image Super-resolution: A Comprehensive Analysis and a New Strategy</td> 
   <td>https://arxiv.org/abs/2004.00448</td> 
   <td>https://github.com/clovaai/cutblur</td> 
  </tr> 
  <tr> 
   <td>EDVR</td> 
   <td>EDVR: Video Restoration with Enhanced Deformable Convolutional Networks</td> 
   <td>https://arxiv.org/abs/1905.02716</td> 
   <td>https://github.com/xinntao/EDVR (PyTorch), <br>https://github.com/xinntao/BasicSR (PyTorch)</td> 
  </tr> 
 </tbody> 
</table>

# 常用数据 #

Set5，Set14，Urban100，BSD100，Sun-Hays 80（均为小数据集）：  
https://github.com/jbhuang0604/SelfExSR

Flickr2K（20G）：  
http://cv.snu.ac.kr/research/EDSR/Flickr2K.tar

DIV2K（8G+）  
https://data.vision.ee.ethz.ch/cvl/DIV2K/