search

omerbt / TokenFlow

Official Pytorch Implementation for "TokenFlow: Consistent Diffusion Features for Consistent Video Editing" presenting "TokenFlow" (ICLR 2024)
https://diffusion-tokenflow.github.io
MIT License
1.56k stars 135 forks source link

Image conditioning #21

Open rakesh-reddy95 opened 1 year ago

rakesh-reddy95 commented 1 year ago

@MichalGeyer Can we use SD v2.1 Unclip architecture to condition on Image and Text to replace the subject with another subject similar to Make-A-Protagonist (https://github.com/HeliosZhao/Make-A-Protagonist)?

MichalGeyer commented 1 year ago

In general, TokenFlow framework can be implemented with any structure-preserving image editing method. So if you have such image-editing method that works on the SD v2.1 Unclip architecture- It should work!

rakesh-reddy95 commented 1 year ago

I have implemented with 2.1 Unclip but the results from the Unet are NULL values.