roey1rg
roey1rg
Thanks a lot! Just wanted to share that I'm making affine augmentations to pictures taken with green screen as a background so it's quite necessary in my scenario.
Thank you for the detailed answer! I tried to run the 512-to-1024 setup you proposed and I faced a CUDA out of memory problems, so I changed the depth multipliers...
Thanks @JensDA, good point
Shouldn't the inverse of r_rot be used to obtain the local positions?
Is there a mirror? thanks