cschenxiang
commented
5 months ago Hello, author. Since I couldn't find your email, I decided to raise the question here. I have downloaded the DID and DDN datasets you provided, as well as the inference results of IDT and DRSformer. I calculated their PSNR and SSIM metrics using the method you provided. The PSNR results are fine, but the SSIM metrics show lower results than those in the paper. So, I wanted to ask if there might be an issue with the datasets provided, or if you think there could be any other possible reasons?
需要注意，对于DID和DDN，需要改动测试数量，1200或1400。 fprintf('psnr=%6.4f, ssim=%6.4f\n',tp/1200,ts/1200) % the number of testing samples DID-Data %fprintf('psnr=%6.4f, ssim=%6.4f\n',tp/1400,ts/1400) % the number of testing samples DDN-Data
Hello, author. Since I couldn't find your email, I decided to raise the question here. I have downloaded the DID and DDN datasets you provided, as well as the inference results of IDT and DRSformer. I calculated their PSNR and SSIM metrics using the method you provided. The PSNR results are fine, but the SSIM metrics show lower results than those in the paper. So, I wanted to ask if there might be an issue with the datasets provided, or if you think there could be any other possible reasons?