This paper proposes a zero-shot text-guided video-to-video translation framework to adapt image models to videos, achieving global style and local texture temporal consistency at a low cost.

https://arxiv.org/abs//2306.07954

YouTube: https://www.youtube.com/@ArxivPapers