Alias-Free GAN

#1

Post by nurpax » Wed, Jun 23, 2021, 4:26 PM UTC

Re: Alias-Free GAN

#2

Post by goldemerald » Wed, Jun 23, 2021, 4:55 PM UTC

After styleGAN-2 came out, I couldn't image what improvements could be made over it. This work is truly impressive.

The comparisons are illuminative: StyleGAN2's mapping of texture to specific pixel location looks very similar to poorly implemented video-game textures. Perhaps future GAN improvements could come from tricks used in non-AI graphic development.

Re: Alias-Free GAN

#3

Post by Gimpei » Wed, Jun 23, 2021, 5:07 PM UTC

Those are some creepy pictures! It's like a photo of the demon inside.

Re: Alias-Free GAN

#4

Post by tyingq » Wed, Jun 23, 2021, 5:10 PM UTC

goldemerald wrote: post #2

After styleGAN-2 came out, I couldn't image what improvements could be made over it. This work is truly impressive. The comparisons are illuminative: StyleGAN2's mapping of texture to specific pixel location looks very similar to poorly implemented video-game textures. Perhaps future GAN improvements could come from tricks used in non-AI graphic development.

>I couldn't image what improvements could be made over it

Still has the telltale of mismatched ears and/or earrings. This seems the most reliable way to recognize them. Well, and the nondescript background.

Re: Alias-Free GAN

#5

Post by minimaxir » Wed, Jun 23, 2021, 5:14 PM UTC

The first two demo videos are interesting examples of using StyleCLIP's global directions to guide an image toward a "smiling face" as noted in that paper with smooth interpolation: https://github.com/orpatashnik/StyleCLIP

I had ran a few chaotic experiments with StyleCLIP a few months ago which would work very well with smooth interpolation: https://minimaxir.com/2021/04/styleclip/

Re: Alias-Free GAN

#6

Post by datameta » Wed, Jun 23, 2021, 5:23 PM UTC

Wow! The rate of progress is truly stunning. I wonder what Refik Anadol could create with this technique.

Re: Alias-Free GAN

#7

Post by benrbray » Wed, Jun 23, 2021, 5:24 PM UTC

Interesting to that this method makes use of Equivariant Neural Networks. Taco Cohen recently published his PhD thesis [1], which combines a dozen or so papers he authored on the topic.

[1]: https://pure.uva.nl/ws/files/60770359/Thesis.pdf

Re: Alias-Free GAN

#8

Post by fogof » Wed, Jun 23, 2021, 5:24 PM UTC

You can see what they're saying about the fixed in place features with the beards in the first video, but StyleGAN gets the teeth symmetry right whereas this work seems to have trouble with it. Why don't the teeth in the StyleGAN slide around like the beard does?

Re: Alias-Free GAN

#9

Post by minimaxir » Wed, Jun 23, 2021, 5:27 PM UTC

fogof wrote: post #8

You can see what they're saying about the fixed in place features with the beards in the first video, but StyleGAN gets the teeth symmetry right whereas this work seems to have trouble with it. Why don't the teeth in the StyleGAN slide around like the beard does?

That's likely the GANSpace/SeFa part of the manipulation.

> In a further test we created two example cinemagraphs that mimic small-scale head movement and facial animation in FFHQ. The geometric head motion was generated as a random latent space walk along hand-picked directions from GANSpace [24] and SeFa [50]. The changes in expression were realized by applying the “global directions” method of StyleCLIP [45], using the prompts “angry face”, “laughing face”, “kissing face”, “sad face”, “singing face”, and “surprised face”. The differences between StyleGAN2 and Alias-Free GAN are again very prominent, with the former displaying jarring sticking of facial hair and skin texture, even under subtle movements

Re: Alias-Free GAN

#10

Post by Geee » Wed, Jun 23, 2021, 5:37 PM UTC

fogof wrote: post #8

You can see what they're saying about the fixed in place features with the beards in the first video, but StyleGAN gets the teeth symmetry right whereas this work seems to have trouble with it. Why don't the teeth in the StyleGAN slide around like the beard does?

In video 9 teeth are sliding.