Alias-Free GAN • Hacker News Board

Re: Alias-Free GAN

#81

Post by jerf » Fri, Jun 25, 2021, 1:22 PM UTC

eru wrote: post #79

jerf wrote: post #14

That's starting to be high enough quality that you could start considering using that for some Hollywood-grade special effects. That beach morph stuff is pretty impressive. Faces, perhaps not quite there yet because we are so hyper-focused on those biologically, but you could make one heck of a drug trip scene or a Doctor Strange-esque scene with much less effort with some of those techniques, effort perhaps even get…

Compare https://news.ycombinator.com/item?id=27559106

First, that's not the same technique and it's not being used for the same purpose.

Second, Hollywood doesn't care about that problem. They will take the best application of the technique, and they don't care if they have to apply a few manual touchups on the result. As long as there is one way of using the system to do the sort of thing they showed in the sample, it won't matter to them that they can't embed a full video game into the neural network itself. They only care about the happy path of the tech.

Someone's probably already starting the company now to use this in special effects, or putting someone on research in an existing company.

Re: Alias-Free GAN

#82

Post by pvillano » Fri, Jun 25, 2021, 6:22 PM UTC

eru wrote: post #78

Lichtso wrote: post #17

The previous approaches learned screen-space-textures for different features and a feature mask to compose them. Now it seems to actually learn the topology lines of the human face [0], as 3D artists would learn them [1] when they study anatomy. It also uses quad grids and even places the edge loops and poles in similar places. [0] https://nvlabs-fi-cdn.nvidia.com/_web/alias-free-gan/img/ali... [1] https://i.pinimg.c…

Yes. It's interesting that imposing what are essentially 2d invariance constraints leads the network to learn what we regard as 3D concepts.

There are some interesting 2d things our eyes do for 3d. If something is on the ground, half is above the horizon and half is below. Parallax is a 2d phenomenon.

Re: Alias-Free GAN

#83

Post by eru » Sun, Jun 27, 2021, 5:43 AM UTC

jerf wrote: post #81

eru wrote: post #79

Earlier quoted context omitted.

Compare https://news.ycombinator.com/item?id=27559106

First, that's not the same technique and it's not being used for the same purpose. Second, Hollywood doesn't care about that problem. They will take the best application of the technique, and they don't care if they have to apply a few manual touchups on the result. As long as there is one way of using the system to do the sort of thing they showed in the sample, it won't matter to them that they can't embed a full v…

> Second, Hollywood doesn't care about that problem.

Hmm, I wasn't trying to nay-say anything here. I mostly agree with your original comment.

See also how in the Gan Theft Auto they are sort-of getting the light reflection for free without having to explicitly teach the network about that parts of physics.