| ▲ | perching_aix 9 hours ago |
| That's kinda what NERFs are (neural radience fields). They actually preceeded this Gaussian story, with Gaussians coming in and outperforming them. Maybe they'll merge later for something even better, I don't know enough about them. |
|
| ▲ | MattCruikshank 9 hours ago | parent | next [-] |
| Sure, but NERFs were trying to match your input photos and poses, not some arbitrary prompt, if I understand correctly. |
| |
| ▲ | Lerc 8 hours ago | parent [-] | | Yes they are image generators. You want image generator generators. A diffusion style process generating gausians instead of pixels. You could possibly do nerfs that way, but it would be effectively generating a trained network. If you managed to do that it would have broad application throughout the field of AI. | | |
| ▲ | dpoloncsak 6 hours ago | parent [-] | | What would this look like in practice? A net that outputs weights for a new net to use? | | |
| ▲ | xigoi 6 hours ago | parent [-] | | Couldn’t you “uncurry” such a process to have only a single network? | | |
| ▲ | dpoloncsak 4 hours ago | parent [-] | | Probably? I'm no expert, just a SysAdmin trying to keep up really... but in my head it's would look like a form of MoE that would gen the 'Expert' model on demand instead of having a variety baked in. That's assuming you could even reasonably train a neural net to output viable weights, of course. |
|
|
|
|
|
| ▲ | cubefox 8 hours ago | parent | prev [-] |
| NERFs have significantly higher image quality than 3D Gaussian Splatting or more recent similar techniques, though they are much slower to render. |
| |
