Is the backbone Denoising Unet need training？

magic-research / magic-animate

[CVPR 2024] MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model

https://showlab.github.io/magicanimate/

BSD 3-Clause "New" or "Revised" License

10.5k stars 1.08k forks source link

Is the backbone Denoising Unet need training？ #142

Closed TZYSJTU closed 10 months ago

TZYSJTU commented 10 months ago

Is the backbone Denoising Unet need to be fine-tuned? Or it is frozen as the SD 1.5 model? Since I only see the Temporal module/Appearence Net/Densepose ControlNet in the given checkpoints, I guess the Denoising Unet need not training?

zcxu-eric commented 10 months ago

yes, we didn't finetune the UNet.