The method of creating kernels
zhiqiangfu opened this issue · comments
I noticed that the function for creating kernel ('anisotropic_gaussian_kernel_matlab') is different from the standard gaussian distribution (e.g. the method that used in IKC, https://github.com/yuanjunchai/IKC/blob/2a846cf1194cd9bace08973d55ecd8fd3179fe48/codes/utils/util.py#L244). I am wondering why a different way is used here. Actually, a test dataset created by IKC with same sigma range seems to have poor performance on MANet, and vice versa.
In addition, I found a code in the 'util.py line 354 shifted_l = l - scale + 1'. I am curious why a shifting process is needed here to make the maximum point away from the center.
Thanks for your questions. We basically follow the implementations of Kernel-GAN, NeurIPS2019, Sec. 5.2 and USRNet, CVPR2020, Sec. 3.1, in which the blur kernel is shifted and the upper-left pixels are kept in downsampling to avoid subpixel misalignments. The generation of blur kernel is identical to Matlab, as note in
Feel free to open it if you have more questions.