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

The difference between v1.2 and v1.1 in the network architecture #81

Closed Ping-Hu closed 1 month ago

Ping-Hu commented 1 month ago

Thank you for releasing v1.2 with support for normal generation! I'm wondering what is the network architecture difference between v1.2 and v1.1? According to the commit history of "pipeline.py" and configs in the directory, diffusers_support, it seems that the customized pipeline is not changed in v1.2. Am I correct? Thank you for your attention!

eliphatfs commented 1 month ago

No, the network architecture didn't change. The data changed.