Open jenkspt opened 1 year ago
That's a great question!
We actually do not apply shading when training in the latent space as it is not clear how it should be defined. Notice however that it can still be applied when fine-tuning our model as explained in RGB Refinement
in Section 3.2. So once you have your base Latent-NeRF you can fine-tune in RGB with the same form of shading used in DreamFusion.
This makes sense. Thanks for the reply!
Thanks for the great research
Seems that diffuse reflectance (a.k.a dot product shading) should work with any number of channels, but I'm wondering what the light color, and ambient light color should be in latent space. Dream Fusion uses light color [.9, .9, .9] and ambient light color [.1, .1, .1] in RGB space. How does this translate to latent space?