Why hasn't anyone used AI to generate new video for the Lost episodes?
Why hasn't anyone used AI to generate new video for the Lost episodes?
We know that AI can be given prompts to generate video. We have many other extant episodes with moving shots of the lead characters's faces to use for training data. We have numerous production still photos from the missing episodes - Marco Polo for instance. We have audio recordings to give us the general gist of what's going on and when, so syncing action to sound should be a matter of specifying what happens and when. Shouldn't it be possible to recreate the lost episodes (or most of them, at least) with the right prompts for a video-generating AI? Isn't this the best possible use for generative AI?
I guess I'm just expressing surprise that no fan has done it yet. We have plenty of fanmade YouTube videos of other franchises, like "Bill Shatner's face deepfaked into Star Trek Continues", "Data de-aged", or "Harrison Ford's face deepfaked into Solo: A Star Wars Story". Is there a reason it can't be done?