Closed Git-H2K closed 7 years ago
Hi, The network performance could be a little different from the CVPR version. We modified the discriminator architecture a little, so it would lead the generator to a different local optima. I believe you got a better result and you are good to go. In case you need the very images at submission time, please send me a mail.
Thanks :)
Hi! I compiled your code and got deblurred results. Next, I calculated PSNR between each deblurred result and sharp image, and averaged them.
However, I got different results.
GOPROSET PSNR (paper, K=1) 28.93 dB PSNR (test, K=1, blurtype & model = blur) 28.49 dB PSNR (test, K=1, blurtype & model = blur_gamma) 29.12 dB
OS : ubuntu 16.04 VGA : titan X
How can I get the same results in your paper ?
Thanks! HH.