Closed blackccpie closed 5 years ago
I am aware of the difference, and that's why I use hard coding. The value is from my original torch implementation.
ok thanks for the tip, the original torch.image native implementation of image.gaussian can be found here, and it gives the sames coefficients as yours.
Hi,
I was just wondering how you generated your hardcoded 7x7 gaussian kernel in img.py. In fact the values differ from the reference formula found in the reference
pose-hg-train
implementation, which is something like:Thanks for your help,
Regards
Albert