那些让人耳目一新的深度学习图像超分辨率重建方法

本文章只记录(按时间),不分析网络结构、优劣。可能会有遗漏,见谅.

1. SRCNN -- ECCV2014

       paper:Learning a Deep Convolutional Network for Image Super-Resolution

      source code: http://mmlab.ie.cuhk.edu.hk/projects/SRCNN.html 

      tensorflow code :https://github.com/tegg89/SRCNN-Tensorflow

2. FSRCNN -- ECCV2016

     paper:Accelerating the Super-Resolution Convolutional Neural Network

     source code:http://mmlab.ie.cuhk.edu.hk/projects/FSRCNN.html

     tensorflow code:https://github.com/yifanw90/FSRCNN-TensorFlow

     pyTorch code:https://github.com/yippp/FSRCNN

 

3. ESPCN -- CVPR2016

     paper:Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network

     tensorflow code:https://github.com/drakelevy/ESPCN-TensorFlow

     pyTorch code:https://github.com/leftthomas/ESPCN

4. VDSR -- CVPR2016

     paper:Accurate Image Super-Resolution Using Very Deep Convolutional Networks

     source code: https://cv.snu.ac.kr/research/VDSR/

     tensorflow code:https://github.com/Jongchan/tensorflow-vdsr

     pyTorch code:https://github.com/twtygqyy/pytorch-vdsr

5. DRCN -- CVPR2016

     paper:Deeply-Recursive Convolutional Network for Image Super-Resolution

     source code:  https://cv.snu.ac.kr/research/DRCN/   

     tensorflow code:https://github.com/jiny2001/deeply-recursive-cnn-tf

     pyTorch code:https://github.com/fungtion/DRCN

6. RED -- NIPS2016

    paper:Image Restoration Using Convolutional Auto-encoders with Symmetric Skip Connections  

    pyTorch code:https://github.com/yjn870/REDNet-pytorch

                             https://github.com/JindongJiang/RedNet

7. DRRN -- CVPR2017

    paper:Image Super-Resolution via Deep Recursive Residual Network

    pyTorch code:https://github.com/jt827859032/DRRN-pytorch

8. LapSRN -- CVPR2017

    paper:Deep Laplacian Pyramid Networks for Fast and Accurate Super-Resolution

    source code:http://vllab.ucmerced.edu/wlai24/LapSRN/    

    tensorflow code:https://github.com/zjuela/LapSRN-tensorflow

    pyTorch code:https://github.com/twtygqyy/pytorch-LapSRN

9. SRDenseNet --  ICCV2017

    paper:Image Super-Resolution Using Dense Skip Connections

    tensorflow code:https://github.com/ppooiiuuyh/SR_SRDenseNet_tensorflow

    python code:https://github.com/wxywhu/SRDenseNet-pytorch

10. SRGAN(SRResNet) -- CVPR2017Perceptual Loss

    paper:Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network

    tensorflow code:https://github.com/tensorlayer/srgan

                                 https://github.com/brade31919/SRGAN-tensorflow

    pyTorch code:https://github.com/leftthomas/SRGAN

                             https://github.com/aitorzip/PyTorch-SRGAN

   Perceptual Loss :

           paper:Perceptual Losses for Real-Time Style Transfer and Super-Resolution

           project:https://cs.stanford.edu/people/jcjohns/eccv16/

11. EDSR --  CVPRW2017(state-of-the-art)

    paper:Enhanced Deep Residual Networks for Single Image Super-Resolution

    tensorflow code:https://github.com/jmiller656/EDSR-Tensorflow

    pyTorch code:https://github.com/thstkdgus35/EDSR-PyTorch

18年之后的就有点多了,这里挑了几个有意思的列举

12. RDN --  CVPR2018

    paper:Residual Dense Network for Image Super-Resolution 

    tensorflow code:https://github.com/hengchuan/RDN-TensorFlow

    pyTorch code:https://github.com/lingtengqiu/RDN-pytorch

13. IDN --  CVPR2018 

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

    tensorflow code:https://github.com/Zheng222/IDN-tensorflow

    pyTorch code:https://github.com/lizhengwei1992/IDN-pytorch

14. DBPN --  CVPR2018

    paper:Deep Back-Projection Networks For Super-Resolution

    tensorflow code:https://github.com/tlokeshkumar/DBPN-tf

    PyTorch code:https://github.com/alterzero/DBPN-Pytorch

15. RCAN --  ECCV2018 

    paper:Image Super-Resolution Using Very Deep Residual Channel Attention Networks

    tensorflow code:https://github.com/kozistr/rcan-tensorflow 

    pyTorch code:https://github.com/yulunzhang/RCAN

2018CVPR转http://openaccess.thecvf.com/CVPR2018.py

2019 CVPR-SR转:http://bbs.cvmart.net/topics/302/cvpr2019paper#11

reference:

1. https://zhuanlan.zhihu.com/p/31664818

2. https://github.com/YapengTian/Single-Image-Super-Resolution

  • 4
    点赞
  • 20
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值