近三年降噪论文整理

一、近两年图像降噪比较好的论文

论文主要是面向真实噪声图像去噪,raw 或 sRGB,方法单帧、多帧。会持续更新,也会写一些论文的阅读分析和复现。

1、Unprocessing Images for Learned Raw Denoising(CVPR2019)

论文链接:https://arxiv.org/abs/1811.11127

项目链接:http://timothybrooks.com/tech/unprocessing/

2、Toward Convolutional Blind Denoising of Real Photographs (CBDnet)

论文链接:https://arxiv.org/abs/1807.04686

代码链接:https://github.com/GuoShi28/CBDNet

3、Deep Iterative Down-Up CNN for Image Denoising (DIDN)

论文链接:http://openaccess.thecvf.com/content_CVPRW_2019/papers/NTIRE/Yu_Deep_Iterative_Down-Up_CNN_for_Image_Denoising_CVPRW_2019_paper.pdf

代码链接:https://github.com/SonghyunYu/DIDN

4、Iterative Residual CNNs for Burst Photography Applications

论文链接:https://arxiv.org/abs/1811.12197

项目主页:https://fkokkinos.github.io/deep_burst/

代码链接:https://github.com/cig-skoltech/burst-cvpr-2019

5、Non-Local Recurrent Network for Image Restoration

论文链接:http://papers.nips.cc/paper/7439-non-local-recurrent-network-for-image-restoration.pdf

代码链接:https://github.com/wenbihan/NLRN

6、Burst Denoising with Kernel Prediction Networks

论文链接:https://arxiv.org/abs/1712.02327

项目链接:https://github.com/google/burst-denoising

代码链接:https://github.com/google/burst-denoising

代码链接2:https://github.com/Pavelrst/DIP_Project

7、Toward a fast and flexible solution for CNN-based image denoising

论文链接:https://ieeexplore.ieee.org/abstract/document/8365806/

项目主页:http://www.ipol.im/pub/art/2019/231/ffdnet-pytorch.zip

代码链接:(MATLAB)https://github.com/cszn/FFDNet

(Pytorch)http://www.ipol.im/pub/art/2019/231/ffdnet-pytorch.zip

8、Learning deep CNN denoiser prior for image restoration

论文链接:http://openaccess.thecvf.com/content_cvpr_2017/papers/Zhang_Learning_Deep_CNN_CVPR_2017_paper.pdf

论文代码:https://github.com/cszn/IRCNN

其他是没有代码的论文

9、Image Blind Denoising With Generative Adversarial Network Based Noise Modeling

论文链接:http://openaccess.thecvf.com/content_cvpr_2018/papers/Chen_Image_Blind_Denoising_CVPR_2018_paper.pdf

10.腾讯优图:Underexposed Photo Enhancement using Deep Illumination Estimation
论文链接:http://jiaya.me/papers/photoenhance_cvpr19.pdf
code链接:https://github.com/wangruixing/DeepUPE

二、视频降噪

1、ViDeNN: Deep Blind Video Denoising

论文链接:https://arxiv.org/abs/1904.10898

code链接:https://github.com/clausmichele/ViDeNN

2、Model-Blind Video Denoising via Frame-To-Frame Training

论文链接:http://openaccess.thecvf.com/content_CVPR_2019/papers/Ehret_Model-Blind_Video_Denoising_via_Frame-To-Frame_Training_CVPR_2019_paper.pdf

3、Deep Graph Laplacian Regularization for Robust Denoising of Real Images

论文链接:http://openaccess.thecvf.com/content_CVPRW_2019/papers/NTIRE/Zeng_Deep_Graph_Laplacian_Regularization_for_Robust_Denoising_of_Real_Images_CVPRW_2019_paper.pdf

4、Learning Deformable Kernels for Image and Video Denoising

论文链接:https://arxiv.org/abs/1904.06903

以往图像降噪

1、Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising

论文链接:http://www4.comp.polyu.edu.hk/~cslzhang/paper/DnCNN.pdf

代码链接:https://github.com/wbhu/DnCNN-tensorflow

其他代码实现:

(keras)https://github.com/husqin/DnCNN-keras

(pytorch)https://github.com/SaoYan/DnCNN-PyTorch

原文链接:https://blog.csdn.net/qq_27871973/article/details/96884653

  • 1
    点赞
  • 21
    收藏
    觉得还不错? 一键收藏
  • 0
    评论

“相关推荐”对你有帮助么?

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
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