Tuesday, February 19th, 2008...5:11 pm - Gary Hayes
Uncanny Valley 2.0 - Faces in Games take a leap
Just spotted at GDC08 this incredible system for bringing very realistic facial animation into games engines and social virtual worlds. It is called MOVA and uses spaced multiple cameras versus motion sensors to map a face actor in real time using a total of 100 000 polygons. But enough about the tech, the proof is in the results. The video embedded above and below is also in the article linked above which has other examples, but they really show how we are entering a space beyond the uncanny valley and moving into the new zone of ‘is this a person or not’ - I don’t want to combine it with my AI post a few days ago cause that starts to get creepy! The real challenge though is rather than pre-rendered CG how to chain together sequences of facial motion data to achieve something that is more seamless - I suppose lots of randomised ‘face is doing nothing’ in-between bits - but too much detail. Be staggered by this video! Steve Perlman (who drove Apple QT and WebTV) now driving MOVA said
“People have never had this kind of data available before in a game context … their heads are spinning,” he said. “What you’re seeing right there is the result of, having time to wrap our heads around this thing and see how we’re going to use it, and yes, we can in fact get a face that looks almost photo-real — you know, not quite, but almost photo-real — running in a game engine today. You can see the difference then between what’s achievable in cinema and what’s achievable right now in video games,” Perlman says. “But next generation game machines, they’ll be able to essentially show in real time what we can do currently in non-real-time using renderers. … Next generation, you’re going to have interactive sequences where people think there’s a live person in the game.”"
Leave a Reply