论文阅读笔记之——《DN-ResNet: Efficient Deep Residual Network for Image Denoising》

本文提出的DN-ResNet，就是a deep convolutional neural network (CNN) consisting of several residual blocks (ResBlocks).感觉有点类似于SRResNet的思路。并且对于训练这个作者所提出的网络，作者还采用了edge-aware loss function

文章focused on denoising either Gaussian or Poisson corrupted images,

在图像去噪中，最常用的是高斯退化模型。the ith observed pixel is

低光散射噪声导致的降级取决于信号，并且通常使用泊松噪声建模

当前的许多去噪网络都是建立LR和HR之间的mapping。而作者认为他们都不适用于实际的图片，由于网络的size。为此作者This training strategy not only allows the resulting DN-ResNet to converge faster, but also allows it to be more computationally efficient than prior art denoising networks.

 

本文的贡献有三：

1、证明了ResNet is effective for image denoising, and using edge-aware loss function signicantly improves the perceptive quality.

2、introduce the depthwise separable ResBlock (DS-ResBlock) to construct DS-DN-ResNet.

3、the proposed DN-ResNet works well for all types of noises, even without knowing the noise level.

 

写本博客之时~其实本人并不在乎作者用什么方法，只是想知道作者是怎么估算noise level的，或者他有没有写到关于noise level的描述