depth control for v1.2 ?

SUDO-AI-3D / zero123plus

Code repository for Zero123++: a Single Image to Consistent Multi-view Diffusion Base Model.

Apache License 2.0

1.56k stars 108 forks source link

depth control for v1.2 ? #71

Open odusseys opened 4 months ago

odusseys commented 4 months ago

Looking at the documentation for version 1.2, I am not seeing a depth/normal controlnet for the initial view generation. If I understand correctly, the normal controlnet only outputs normals instead of images in this version. Does the old v1 depth controlnet still work with the new model ? Thanks

eliphatfs commented 3 months ago

The v1 controlnet will possibly not work with the new model but you could have a try. Our guess is that the performance of v1.2 will not be much higher than v1.1 if conditioned on geometry but we may add v1.2 controlnets for more control modalities later.