In January I went with a friend to see Avatar, the much-acclaimed film by James Cameron. Its claim to fame was the near-monopoly of computer generation in the making of the film, coupled with the fact that viewers could watch it in "3-D." (And this, of course, meant that it is the most expensive film to date--a title I'm sure it will hold for at least another three months.)
The 3-D part was cool, not only because it's always cool to see depth on the screen, but because the film knew when to stop. 3-D films sometimes succumb to the temptation to send a sharp stick flying toward your eye in every scene. Like a five-year-old who wants to keep making that sound even though it's not funny anymore, these films overuse the ruse of making something "jump out of the screen" at an audience who has not, in fact, come to get whiplash.
Avatar was not seduced by this lure. Instead, with only a few near-miss reminders that you're actually in the film, the "3-D-Wow!" effects made way on the stage for normal encounters with depth. The plot and characters were given room to do their Aristotelian thing.
The part of this film that really impressed me, however, did not strike me until the film ended. And it struck me in part because I only realized it as the lights were coming up. The computer-generated faces in the film had such a humanity to them that my viewing mind accepted them as real.
The human body is difficult to replicate on the screen. I've played a lot of video games that shoot for realistic representations of people, so I've seen many attempts to convince the viewer that the CG virtuality is reality. But humans--and all organisms, in fact--are tricky things. They're always in motion. Even when a person is standing "still," tiny movements betray life, thought, and unconscious presence. If these are absent, as is so often the case in films and video games, the viewing mind perceives that the image is unreal. And soon we tire of the mockery.
This problem comes to a head with faces. So many tiny muscles make the face difficult to replicate for a given emotive expression, let alone over seconds of passive, unconscious movements. Even the sleeping face is never quite still, in the way a photograph is still. Thus computer generated faces, no matter how detailed in texture, usually appear wooden. The viewing mind registers them as fake, even when it might otherwise be taken up by the plot or dialog on the screen.
Avatar made a huge leap in presenting the face as an extension, the quintessential emblem, of a living being. As I watched the film I became engrossed in the action, character development, and narrative precisely because I was not distracted by the faux finish of the characters' faces. I don't know anything about computer graphics and CG animation, but from a viewer's perspective I have to say: James Cameron and his teams did a great job in Avatar. They've put a whole new face on CG animation.