CGI films and the uncanny valleyPosted: June 24, 2011
There’s been a lot of CGI films over the past decade. You have dedicated studios such as Pixar that make them. But then there’s also the hyper-realistic types of films as well. I remember being astounded by the computer generated imagery of Final Fantasy the Spirits Within. It was an amazing film in terms of graphics, but it was often panned for its story.
But the closer we get to realism, the freakier CGI can become (hence the uncanny valley). Of course, with new technologies available, the more realistic we can make things, and we’ve seen that with live action integration with animated computer generated imagery. It still occasionally alludes us, this sense of realism though, mostly because it tries to be (but not quite makes it) realistic – some bits can be lifelike, other bits make you realise it’s just an illusion.
Things like art style also play a factor. Final Fantasy VII Advent Children for example had amazing graphics (also impossible physics, but heck it’s sci-fi/fantasy). The reason uncanny valley didn’t crop up then was because a, the characters were established as sprites and not as real and b, the art style meant that the faces, eyes, structure etc… were not realistic.
But how do you get realism in computer graphics? Have a gander at this video:
But even though we have the possibility at least of hyper-realistic computer generated characters, do we still empathise with them in the same way? Well, I personally think we empathise more with cartoon, stylised and misshapen characters because they’re accentuated in certain respects. And why are even their performances are accentuated? Because it’s not going for realism. Here’s what I mean:
Take a look at this shot. It’s from the Final Flight of the Osiris. It had hyper-realistic graphics. Every movement was based upon what real actors do, with the artwork and movements imitating as close to reality as possible. Meanwhile:
Toy Story’s style not only necessitated that it be different from reality, but that freedom meant that every part of the image, from body language, colouration, expressions are all purposeful. What do I mean by that?
It basically boils down to expression of the character. With hyper-realism we have to know that the world is realistic, that the character lives inside that world. Without hyper-realism we instead concentrate on the character’s feelings, reactions and motivations. The thing that draws us in with a pixar film (for example) are the characters. With other hyper-realistic animations, it’s actually the story and the world.
That’s not to say that hyper-realism is bad because of it. I appreciate the effort to get to hyper-realism and find it a unique storytelling tool. It even has the advantage over cartoon style CG which is that it can be placed into live action films without too much hassle.
I think sometimes we forget that film isn’t just trying to be one thing. All art forms are dynamic and we should appreciate all the beauty in films.