- cross-posted to:
- [email protected]
- [email protected]
- cross-posted to:
- [email protected]
- [email protected]
There is a discussion on Hacker News, but feel free to comment here as well.
There is a discussion on Hacker News, but feel free to comment here as well.
Everyone mentions LOTR, but they did those by duplicating the images of the real actors, and done at such distances no discernable detail is visible.
This shit just looks awful. Regardless of what opinions anyone has on AI vs real people, nobody wants it to look ugly. This is ugly AF.
" “Prom Pact,” a B-grade teen movie on the Disney Channel."
I don’t think this is comparable to LotR, this isn’t a blockbuster movie it’s a trashy disposable low budget teen movie.
These CG extras are probably cheaper per unit than the old “stick two rows of humans in front of a dozen rows of cardboard cutouts” trick they’ve been using for decades.
You’re not correct.
https://www.cnet.com/culture/entertainment/features/how-lord-of-the-rings-used-ai-to-change-big-screen-battles-forever/
You are not understanding the difference between using AI to create the movement of the crowd, and using AI to digitize entirely new actors. They digitally duplicated the actors, and using a more video game like AI created the motion of those actors in the scene.
They talk in length about this in many commentary tracks of the actual releases of the movies.
No, I understand. You don’t understand how any of the technology works so when you listened to the commentaries you misunderstood.
Here’s one example of Weta showing off Massive used for one of many shots at Helm’s Deep.
Are you being voluntarily obtuse and not recognizing the fact that the general public means “machine learning” when they say “AI”, or are you saying that massive uses machine learning ? Because as far as I know, at least when this movie was being made, massive was just “intelligent agents” (no machine learning involved) + softbody physics. The end result is a crowd that looks realistic enough from afar.
No I am not disputing the use of intelligent agents in Massive. In this discussion I’m the person who namedropped the tool. What I’m disputing is the interpretation of what WETA did to produce the large scale combat. People have misinterpreted that some mid range scenes used judiciously copy pasted actors and props to mean that’s all they used when clearly Massive was used with entirely digital 3D modeled and hand textured actors and props for longer distance shots when they needed to work at scale.
Massive, like its predecessor “a flock of birds” is a very very useful particle control and generation system and it was definitely in use here with hand modelled and textured characters and props.
Here is an alternative Piped link(s):
Weta showing off Massive
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I’m open-source; check me out at GitHub.