Greetings Axeheads and welcome back! They say everything old is new again and for once they’re not talking about annoying facets of popular media and culture. As you’re probably aware, there’s a lot of common ground shared by Hollywood visual fx studios and the game industry. This is particularly true where motion or performance capture is concerned.
As game developers attempt to create more and more realistic characters they rely more and more on the techniques for reproducing realistic movement that have been pioneered by the film industry. However, as this technology evolves artists are finding it can be used not just to capture the motion of the human body or face, but to reverse (or hasten) the effects of time and aging on an actor; maybe even to raise the dead.
I give you a spoiler warning now; we’re going to be talking about major plot points from several recent and upcoming films. If you don’t want to know anything, turn back now.
The first time I became aware of anti-aging fx being employed by CG artists was X-Men 3: The Last Stand. The fact that it’s a horrible film aside, towards the beginning there’s a scene where younger incarnations of Professor X and Magneto visit a teenage Jean Grey. The actors were shot without special makeup or any sort of consideration for the visual fx that would be applied in post production.
Now, this computer-generated-zombie-botox technology was in its infancy then (2006) and the results weren’t too stellar but as a baseline, things could only get better.
And get better they did. For the 2008 film The Curious Case of Benjamin Button, visual fx house Digital Domain was tasked with creating a computer-generated head for actor Brad Pitt. Not only would this CG creation need to be completely photo-realistic, but would also have to accurately portray Pitt as a man of 60, 70, and 80 years old. As Ed Ulbrich explained during his appearance at TED, Digital Domain had to abandon the existing paradigms of motion and performance capture and develop an entirely new approach.
I know this is a really long vid, but I promise you its worth watching. There’s some really fascinating things discussed here.
Part of what makes creating believable digital characters so difficult is a phenomenon referred to as the Uncanny Valley. Essentially its a thesis which (in part) proposes that people can readily empathize and bond with characters who have human-like features (think cartoon characters). Our brains know its not a real person but because of the stylized rendering, we forgive the lack of realistic behavior. However, the theory states that there is a point in this evolution where a character looks so life-like that our brains expect all the nuances of interacting with other human beings and when that’s not present, we find the effect repulsive.
We’re only interested in the first forty seconds here; don’t feel like you need to watch the whole clip.
In 2009, Legacy Effects created a nostalgic digital cameo for Arnold Schwarzenegger in Terminator Salvation. Unlike Benjamin Button, the fx house did not have access to the govenator and had to create their CG model based on a life-cast taken of teh Austrian Oak during the original Terminator. The effect is good, but there’s something about it that sets off the “fake!” alert in our minds. The vfx artists failed to cross the Uncanny Valley; there is some subtle quality that our brains are looking for, but failing to see.
Now, its 2010 and Digital Domain is rolling out the next generation of their emotion capture technology in the upcoming Tron: Legacy. Jeff Bridges will once again have duel roles in the film as Kevin Flynn and Flynn’s software creation Clu. For the character of Clu, the film makers decided they wanted to see Bridges as a young man. In the first actual trailer for the film, we get a glimpse of Digital Domain’s work.
There are two shots of a de-aged Jeff Bridges as Clu at 1:37.
Years ago, when CGI was first being employed in movies like Terminator 2 and Jurassic Park, I can remember people speculating that one day the technology could be used to bring actors back from the grave. For the first time, I’m inclined to agree with them. Forbes ran an excellent piece last month discussing this technology, the Uncanny Valley, and the implications as it continues to evlove. The author postulated about a new Indiana Jones movie with a young Harrison Ford battling evil-doers along side Humphrey Bogart reprising his role as Rick Blaine from Casablanca. I must admit its an intriguing thought. More to the point, even a year or two ago it might’ve been considered impossible; that’s clearly no longer the case.
So what about gaming? Many of the techniques of the film industry trickle down to game devs especially, as I pointed out previously, performance capture. How might game developers employ something similar to Digital Domain’s emotion capture in the future? How many console or graphics card generations are we from seeing photo-realisitc characters rendered in real-time? Where do the lines begin to blur between what we traditionally think of as games and the emerging interactive experience paradigm? How far away are we from a fusion of film and gaming where two people can watch a movie and have two entirely different experiences based on their conscious and unconscious input? Are we seeing the beginnings of that even now with titles like Heavy Rain? Will Tron 3 and Indiana Jones 6 have dual analog nubs?
Possibilities. Endless possibilities.











I think that as time goes on, we’ll see better graphics and ways to interact, from the movie industry as well as from the gaming industry in the next decade or two.
Although something tells me what we know as gaming might evolve in to something completely different in the future. Recently I saw my 3 year old nice kicking ass at this touch screen game in my iphone, making videos and playing sword fight in wii sports resort(not everything at the same time, although that would be awesome)
Think for a second how was gaming when you were 3 and you can see the humongous advance of technology and how it is appreciated this days. This might give us a good idea of what future has to offer, but also it leads me to believe that gaming as we know will evolve in to something that we as “grandpas” might think that is “too crazy” or “not fun” it will be funny to see the end result in the future.
Motion capture in controling games… yes well… we will see. Natal/MOVE are the ones to bring it with the cameras… No jelly to rub your face with, bummer.
Digi-hair is going to rock the socks off little girls and girly boys(who?). So it’s a shame they didn’t discuss that in the first vid. Since that’s the niche(?) that’s going to bring in the funding in this whole “type of deal”.
I hope they won’t bring back the dead… too many wanna-be’s will have broken hearts. And it’s not that healthy to remanise about the “good old days”… IMHO. As a TECH demo it’s okay.
Devs got more tools to play with… doesn’t mean they are going to or that it will be any good. I think motion capture animation in GAMES hasn’t been that big of a deal. And that’s been around for some time now.
Well, it will be intresting, not going to hold my breath tho.
[...] week in discussing the state of life-like computer generated characters, I used a Ted talk from Ed Ulbrich to illustrate the current state of that technology. This week, [...]