search

showlab / Tune-A-Video

[ICCV 2023] Tune-A-Video: One-Shot Tuning of Image Diffusion Models for Text-to-Video Generation
https://tuneavideo.github.io
Apache License 2.0
4.25k stars 386 forks source link

Image+Text guided video editing #70

Open rakesh-reddy95 opened 1 year ago

rakesh-reddy95 commented 1 year ago

@zhangjiewu Currently tune-a-video is a text guided video editing. May I know how we can apply this for Text+Image guided video editing. I just want to replace the subject in a video from the given photo and give the text prompt for action? Can you please provide if you come across in your research?

plzhai commented 1 year ago

i am also focusing on this, maybe we need to selelct the specifc noise in the begining of the inference process. Please let each other know if there is an answer.

rakesh-reddy95 commented 1 year ago

You can refer to make-a-protagonist (https://arxiv.org/abs/2305.08850).