3 ms·
Not to be a skeptic or anything, but how do we know normal maps etc weren’t enriched into the datasets by the image gen companies? I understand this paper link
by vslira 3y ago
Not to be a skeptic or anything, but how do we know normal maps etc weren’t enriched into the datasets by the image gen companies?
I understand this paper links to open source model where that can be verified, but maybe this is one secret sauce of these more advanced models?
- dougmwne 3y agoYou would need to train on pairings of normal map images to source images. To my knowledge, that’s not a common training technique and this ability seems to bridge across several open models.
- coffeebeqn 3y agoIt’s also kind of cool if it “understands” a normal map and that it maps to specific 3D geometry. People generally use 3D tools like Zbrush to sculpt the details to normals. Probably some people can draw them too