What is the most modern technique for generating a video from an image+driver video?
NVIDEA's Maxine and Live Portrait microservices look very promising but they are still in early access. But if you dig into Live Portrait it seems to be built off of a three year old github repo. Example from the repo: https://i.redd…