Dear Wired,
Oh you. How did you know I had planned to write a post on this exact topic?
This demo of performance capture technology comes from game developer Quantic Dream, known for implementing motion capture into past video games. For those of you who are unfamiliar with the term, performance capture involves animating a digitally-constructed 3D character with the facial and body motions of a real actor or actress, through the detection of specially-placed dot markers on the person.
So why use this?
For one thing, it’s probably a easier to work with. If you’ve played QWOP before, you’ll get a sense of why the mechanics behind muscle movement are deceptively challenging. You don’t think about controlling a smile; it just happens.
But the more important reason is that performance capture gives animated characters more believable emotion and movement.
One of the main goals of 3D animation is to try and bridge what’s known as the “uncanny valley.” This is that semi-disturbing twilight zone where the viewer is conscious that they’re watching a fake human, essentially; it’s that gap between an intentionally cartoonish style and flawlessly realistic rendering. Movies like Avatar, Planet of the Apes, and Lord of the Rings have all attempted this, with varying levels of success.
It is especially fascinating what Avatar has done to revolutionize this technology. When you’re watching the movie, you’re watching a real performance, but one that has been mediated and augmented by the James Cameron’s lens.
However, the couple of times I watched the movie, I know I wasn’t thinking at all about how the Na’vi weren’t real. I was completely immersed in the world of Pandora.
Let’s go back to the animation at the top of the post and compare. This animation is similarly at the forefront of performance capture technology, but if you were like me, you very distinctly felt that Kara had not yet crossed the uncanny valley.
I think the reason Avatar succeeds where Kara falls short is that the Na’vi are at least a level removed from humans, and we don’t expect quite the same level of perfection, partially because… well, we don’t have any prior expectations for aliens.
At the same time though, consider this: do you actually want animated humans to be indistinguishable from real actors?
There are so many issues that stem from this, considering the intrinsic artistic qualities for both live-action and performance capture, the possibility of dubious ethical situations (like doctoring videos of individuals), and the effect this would have on the acting profession itself… plus, I’m sure a sizable portion of viewers just find characters animated with motion capture downright creepy.
Andy Serkis, in the Planet of the Apes.
Additionally, how does this change our notions of what animation is? Should there be a new genre outside of animation for these? How much of it comes from the actors, and how much from the animators?
Will we expect animated movies to be mere shells for real actors to fill?
It remains unclear whether performance capture is the future of animation, but it will most certainly be in the future of animation. If you’re unhappy about it… well, I’m sure we can move around a few capture points and fix that right up.
Leave a Comment so far
Leave a comment
pixelpaints 