Would it be feasible to provide more than one view angle as input to better the quality of the outputs.

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

Would it be feasible to provide more than one view angle as input to better the quality of the outputs. #75

Open TiasheKnight opened 2 months ago

TiasheKnight commented 2 months ago

Would it be feasible to set it up so that I can provide more than one view angle as input to better the quality of the outputs. I have front, back and side views of an animated character. Perhaps similar to how its done in Photogrammetry?