yaseryacoob

I am missing something, I tried an inverted latent, 18x512, attached, and the odd thing is that the size of M.dlatents[0] is 1,512 where it is 2000,512 if I use...

Dear Alex, This solved it, it didn't occur to me until you mentioned it. I again thanks you for sharing the code and the effort you put into it. So...

I figured you must have seen these, and I fully understand the complexity of research code and also the challenge. I actually mean to be encouraging of the approach and...

Actually research and engineering blend when significant architecture changes are made. The changes you made can and should lead to different outcomes from StyleGAN2, but there are a number of...

Yeah I forgot about the restyle approach! I had discussed things with them last week. You can see the thread there. We are missing the last 5-10% in quality, and...

@ganjbakhshali Can you explain how you did it? PTI fine tunes the generator, so in effect you one has to put the custom generator in the styleflow pipeline.

Since you actually went through effort of evaluating it, even if it didn't improve, there are insights that can develop from the new framework on such data. Your work so...

Thanks for the explanation. I noticed less diversity (than I expected) when I generated 1K images (but using a fixed truncation after it was computed in the block before-last in...

We have used a number of approaches, a quick summary 1. Similar to STYLEGAN2-ADA projection using VGG metric. 2. VGG+L2+LPIPS+ID_loss (similar to pSp). For now we working with FFHQ 256...

Thanks for the effort, unfortunately I am running on 16.04 and it does not build, I get errors such as /usr/include/c++/4.8.2/bits/c++0x_warning.h:32:2: error: #error This file requires compiler and library support...