Open Delicious-Bitter-Melon opened 3 months ago
In some cases, the wrong computation manner will increase PSNR but in some cases it decreases PSNR.
Besides, I am certain that the FID calculation method used by Disco is correct, but I am not sure if other metrics such as FVD are correct.
The Disco code of computing PSNR as follows is wrong: https://github.com/Wangt-CN/DisCo/blob/8538889c9ee9edd8dd43ffee182d1a91ce7a9828/tool/metrics/ssim_l1_lpips_psnr.py#L13.
As pointed out in https://github.com/Wangt-CN/DisCo/issues/86, the accurate code is mse = np.mean((original/1.0 - compressed/1.0) 2) instead of mse = np.mean((original - compressed) 2) , because original and compressed images are uint8 in their code, and (original - compressed) 2 will cause numerical overflow.
If you use their evaluation code of computing PSNR, please update your results.