search

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

something about trainning detail.. #69

Closed MouseChannel closed 3 months ago

MouseChannel commented 4 months ago

how to embedding Camera_pos(elevation,azimuth) into model? so I can fine tune it for my own task

YuxuanSnow commented 4 months ago

The camera poses are fixed, thus no camera poses are injected into the diffusion process.

eliphatfs commented 3 months ago

Thank you for answering the question @YuxuanSnow !