Will there be a usable model or pipeline in Diffusers?

williamyang1991 / Rerender_A_Video

[SIGGRAPH Asia 2023] Rerender A Video: Zero-Shot Text-Guided Video-to-Video Translation

https://www.mmlab-ntu.com/project/rerender/

Other

2.97k stars 202 forks source link

Will there be a usable model or pipeline in Diffusers? #118

Closed sunhaozhepy closed 11 months ago

sunhaozhepy commented 11 months ago

Hi,

Thanks for your great work! I read on the project's homepage on GitHub that there will be a Diffusers implementation. May I ask how are things going, and when will the implementation be released? Instead of using the normal ControlNet pipeline, I'm willing to combine ControlNet and img2img to apply your model to my task, and I think it's easier to customize the model with the Diffusers package.

williamyang1991 commented 11 months ago

The diffuser code is almost done. @SingleZombie maybe ready in this month?

sunhaozhepy commented 11 months ago

That is exciting news. Can't wait for the release!