Thread

1. Lerc · 1790845274 · focus · HN ↗

   I'm rather surprised that it doesn't take the z-buffer as an input. I would have thought that would have provided useful information, it's one of the more useful forms of contolnet.

   1. strangecasts · 1790846395 · focus · HN ↗

      I think this is mainly so it can use the existing hooks for DLSS upscaling without requiring changes to the renderer, AMD is working on a comparable method which uses adapter networks to slot normals and material properties from the renderer into the diffusion model: <a href="https:&#x2F;&#x2F;gpuopen.com&#x2F;learn&#x2F;temporally-stable-generative-illumination&#x2F;" rel="nofollow">https:&#x2F;&#x2F;gpuopen.com&#x2F;learn&#x2F;temporally-stable-generative-illum...

Unofficial Hacker News client; not affiliated with Y Combinator.