cszn
commented
4 years ago DPSR and USRNet are similar but essentially different methods.
- DPSR is a plug-and-play method which only needs to train on bicubic degradation.
- USRNet is a deep unfolding method which can be trained in an end-to-end manner.
I will release the code after the sponsor's approval.
The D module takes the noise level as input, you can refer to FFDNet as an example. It is not necessarily the same as ResUNet.
You may try to implement it by yourself.
wangqs97
Hi cszn, thanks for this fantastic paper. I think the difference between this paper and DPSR is that H module and D module, but after reading I still don't know the details of D module.
Would you release the traing code in the future? If not, could you elaborate on the implementation of the D module? Thanks in advance.