Hacker News Reader

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.

View on HN

>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.

Mismatched reflections across eyes is the dead give-away for me.

Teeth too. Partially covered objects in 3D space have been hard for a GAN to figure out. (See also hands)

I wonder what dataset you could even use to tell a GAN about human internals. 3D renders of a skull with various layers removed?

I've noticed the same thing with ESRGAN -- teeth are always awful. I'm looking forward to the day when someone figured out how to fix that; I have a few sentimental images taken with a cell phone I would love to see upscaled and cleaned up.

Reply on news.ycombinator.com