pwp1208/Domain_Translation_Multi-weather_Restoration

About the paper experimental data

Closed this issue · 8 comments

Hey Bro:
I want to reproduce the experimental data of your paper, may I ask your NIQE and other results, is the input dataset picture adjusted to 256*256 test?
image

Hi,
While testing all the images are resized to 2562563.

ok,bro

Hello, I tested the PSNR of the 492 pairs of pictures in the SOTS_outdoor according to your paper method, and I found that the measured PSNR is less than the value given in your paper?What is the reason for this?I first converted all the pictures input to the model into 2562563 format, and then output a clean image through the model you posted, and finally tested the PSNR and SSIM between the output image and the label image, I would be grateful if you could give me guidance!
1714217357535
1714217375207

Hi,
We used 500 outdoor images for testing in .png format.
Also, we calculated all the parameters on MATLAB.
Hopefully with these changes you will get the appropriate results.

thank you,thank you very much

Hello authors, I have some obstacles in recovering PSNR and SSIM, can you share the MATLAB code of your evaluation with me? I would appreciate it if you could share it with me! my gmail :yesimwlt@gmail.com

I hope you have received it.

of course. Thank you very much.