Will something like Sora 2 actually be used in Hollywood productions? If so, what types of scenes?
I imagine it won’t necessarily be used in long scenes with subtle body language, etc involved. But maybe it’ll be used in other types of scenes?
Will something like Sora 2 actually be used in Hollywood productions? If so, what types of scenes?
I imagine it won’t necessarily be used in long scenes with subtle body language, etc involved. But maybe it’ll be used in other types of scenes?
I saw a famous actor-director (can't remember who, but an A-list guy) said it would be super valuable even if you only use it for establishing shots.
Like you have an exterior shot of a cabin, the surrounding environment, etc — all generated. Then you jump inside which can be shot on a traditional set in a studio.
Getting that establishing shot in real life might cost $30K to find a location, get the crew there, etc. Huge boon to indie films on a budget, but being able to endlessly tweak the shot is valuable even for productions that could afford to do it IRL.
Probably Ben Affleck. https://www.youtube.com/watch?v=ypURoMU3P3U
Searched around and found it. It was actually Ashton Kutcher's interview with Eric Schmidt.
Kutcher mentions the establishing shots, and I'd forgotten also points out the utility for relatively short stunt sequences.
> Why would you go out and shoot an establishing shot of a house in a television show when you could just create the establishing shot for $100? To go out and shoot it would cost you thousands of dollars.
> Action scenes of me jumping off of this building, you don’t have to have a stunt person go do it, you could just go do it [with AI].
Casey Affleck is currently shooting a horror vampire period piece using Comfy UI and an Unreal Engine Volume. The AI is used for the background plates. It's just a test, but it's happening right now.
Jason Blum is also getting really into the tech.
Wow. What an intelligent take. I would have never expected this from Ben Affleck. He seems extremely familiar with the technology and it's capabilities and limits.
I think it will mainly be used for inpainting and outpainting, i.e. for adding and removing stuff in a scene. Things that currently have to be done with relatively expensive CGI. More complex things, especially things that have to look identical between scenes, or have to look a very specific way, will still require filming or classical CGI, or both. (For now.)
People use these for sure but the biggest problem with these I feel is that they produce ”finished shots” with 8 bit colors and heavy grading. It’s hard to mix it with the other material which actually looks quite bland while it is being worked on. Would be great if somebody would train a model on raw footage.