Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

class PReNet(nn.Module): class PReNet_r(nn.Module): class PRN(nn.Module): class PRN_r(nn.Module): #3

Open
jilner opened this issue May 4, 2019 · 8 comments

Comments

@jilner
Copy link

jilner commented May 4, 2019

作者你好,在network.py中上述四个网络结构,为什么输出的时候倒数第三行的结果要加上input呢?x = x + input,这个应该是已经derain后的结果,在加上输入的图,是不是多余了?

@csdwren
Copy link
Owner

csdwren commented May 5, 2019

这是用了residual learning,网络学的是negative rain streak layer。直接输出x 不加input也可以,数据集指标稍降一点,视觉上没区别。

@jilner
Copy link
Author

jilner commented May 5, 2019

谢谢 再麻烦问下 negative rain streak layer 是哪篇文章呢?

@csdwren
Copy link
Owner

csdwren commented May 5, 2019

目前几乎所有的去雨网络都用了residual learning,DDN是第一个用在去雨的。DnCNN第一个把这个策略用到去噪 https://github.com/cszn/DnCNN

@jilner
Copy link
Author

jilner commented May 5, 2019

以前只注意到中间的残差会加上之前的输入 但是好像第一次见到在最后输出的时候还要加上最开始的输入

@Rocky1ee
Copy link

这是用了residual learning,网络学的是negative rain streak layer。直接输出x 不加input也可以,数据集指标稍降一点,视觉上没区别。

这是用了residual learning,网络学的是negative rain streak layer。直接输出x 不加input也可以,数据集指标稍降一点,视觉上没区

您好,有个小疑问:
x 为 negative rain streak,input + x = deraining image。
要是不进行此操作,直接输出x如何得到deraining image呢?

@csdwren
Copy link
Owner

csdwren commented Mar 15, 2020

请看networks.py中的 class PReNet_LSTM

@Rocky1ee
Copy link

请看networks.py中的 class PReNet_LSTM
image

@csdwren
Copy link
Owner

csdwren commented Mar 15, 2020

是的

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

3 participants