Image Metrics May Revolutionize Facial Animation

iStorm writes, "I've been interested in computer animation for a long time and have recently started cracking down on my studies in an effort to eventually move myself from hobbyist to professional... then I find this article about Image Metrics, which can map an actor's emoting onto a generated face or onto the image of another actor, living or dead. How does a seasoned animator view this sort of push ahead in technology? If so much of the creative process is made so easy, where's the need for traditional animators spending exponentially larger amounts of time to create work of equal or lesser quality? How did animators view motion capture when it first appeared? Will there still be room for creativity if this tech comes to fruition?" The article doesn't say what kind of time or processing power Image Metrics's "high-fidelity, performance-driven facial animation" requires.

Comments

There is always art in animation

[Anonymous Coward]

I doubt that this technique will knock animators or traditional animation out of business. Animation is art. Did the video camera kill painting? Did the internet kill reading?

Animation from an animator gives it style, and feeling just as much as an actor does. Just watch any old Disney cartoon if you want to see the flow of such animation.

Re:There is always art in animation

[blahplusplus]

"Animation from an animator gives it style, and feeling just as much as an actor does. Just watch any old Disney cartoon if you want to see the flow of such animation."

Another issue is that much animation we see is not very realistic to begin with, it does not obey the laws of physics. Most animation is exagerated to look good and for visual, visceral and emotional impact, in fact most animation is not real. If you look at the real world many real world animated things are in fact... boring. Some of the most entertaining and hilarious animations are those that are in fact not real to begin with.

The the things that give superheroes in cartoons or animated series their "mmmph" is the fact that you get unreal exagerated actions and reactions that make them interesting to begin with.

Not a new question...

[The Dalex]

How did painters view photographic technology when it first appeared?

Re:Not a new question...

[EaglemanBSA]

Picasso said "I have discovered photography. Now I can kill myself. I have nothing else to learn." Painting and other forms of art, however, are not dead. Keep in mind that for Finding Nemo, artists actually strived to make such entities as the whale, etc. less real, for the sake of the style the wanted to show. Animation/art isn't always about making something seem real.

Not to worry

[corroncho]

Crappy acting will still be crappy acting. Just like the foley artist is still happily employed enhancing the audio soundtrack (either digitally or old fashined foot stomping). The animator will remain gainfully employed improving and enhancing the final product.
___________________________
Free iPods? Its legit and simple. 5 of my friends got theirs. Get yours here!

Rendering Time

[Telvin_3d]

The article doesn't say what kind of time or processing power Image Metrics's "high-fidelity, performance-driven facial animation" requires

I don't care how much processing power it takes, unless we are talking simulation on the level of some of the whole-world-weather simulations any additional processing will be a drop in the bucket compared to the current amount of time and processing power already devoted to any production quality animation.

The evolution

[eebra82]

How does a seasoned animator view this sort of push ahead in technology? If so much of the creative process is made so easy, where's the need for traditional animators spending exponentially larger amounts of time to create work of equal or lesser quality?

Technology usually advances so that it is not only more advanced, but also more efficient. It's fairly obvious that Hollywood studios (just an example) would want cheap CGI, and since there's a need for this to happen, there's also someone working on making that happen.

A skilled animator shouldn't be worried, however. Creativity is hard to replace with software and someone will always have to create whatever's portraited. How it's done and how fast is a different question.

This seasoned animator's view

[tinrobot]

I've been doing 3D character animation for well over a decade, and I've also been exposed to automated facial animation systems including mocap (in many of it's various forms) over the years. I actually think mocap is not bad for certain applications, particularly whole body stuff like athletics. If you really want that golf swing to look like Tiger Woods in his video game, then mocapping him is a very valid option.

What it's not good at, however, is animating the face. People have been trained since birth to observe human faces and we're experts. It makes us very aware of anything that's unnatural. Only a human who innately understands the subtleties of human emotions can truly finesse facial animation so it looks pleasing to the human eye. An animator is just that type of person. We study facial expression, musculature and all sorts of things, then combine it with acting skills and artistic knowledge to make a result that's looks pleasing to the eye (or not.. depending on budgets and deadlines - and I suspect this technology will filter down to the low end productions that don't care as much about the final results)

Re:This seasoned animator's view

[nasch]

Only a human who innately understands the subtleties of human emotions can truly finesse facial animation so it looks pleasing to the human eye.

You're saying this technology could never become good enough to copy a human face well enough to be indistinguishable to a human observer? Or are you saying something else?

Re:This seasoned animator's view

[ConceptJunkie]

If he is saying that it couldn't fool a human, I would agree. While I've seen static images that are startingly realistic-looking, I have yet to see animation that gets anywhere near the far edge of the Uncanny Valley.

That's why we will continue, for a long time, to see animation like The Incredibles, who don't look even the slightest real, but have very convincing expressions and poses.

People will keep trying, and eventually succeed to duplicate photorealistic animated human expression, but I give it 10 years or more until that happens.

Technology always moves the curve...

[maillemaker]

>If so much of the creative process is made so easy, where's the need for traditional animators
>spending exponentially larger amounts of time to create work of equal or lesser quality? How did animators
>view motion capture when it first appeared? Will there still be room for creativity if this tech comes to fruition?"

This sort of thing has always come along. For example, WYSIWYG applications like Front Page let people like me create web pages without knowing hardly any HTML. Suddenly /anyone/ could make a pretty good looking web page. Did a lot of web page authors bite the dust then? Sure. But the ones who remained advanced their art - now they are masters not just of HTML but PHP, AJAX, JAVA, Flash, and a host of other cutting-edge web functionalities.

Computer Aided Drafting did the same thing to mechanical drafters who worked on drawing boards with pencil and paper.

That's kind of the point of technology - to make what were once difficult or tedious expenditures of effort become effortless. Talented people who specialized in those old efforts will have to move on to tackle new things that are still difficult. There's always a new cutting edge.

Steve

Exaggeration

[Anonymous Coward]

Most good animations exaggerate the motions of the characters to add life and energy. On screen, having everything turned up a notch above what it would be in the real world helps convey the message you're going for. And of course facial expressions are included in all of that. Using mocap only captures reality(or at least it tries to) and doesn't allow for any freedom to exaggerate things as the animator sees fit. And even if you wanted to start with mocap and build in some other motion from there, it makes it difficult to do so because of all the noisy data that it collects. Picking through that and cleaning it up so you can work with it can be more time-consuming than just doing it from scratch yourself.

So no, I don't think it's going to make such a big impact that animators will have to worry about keeping their jobs. Humans are actually good at some things!

Re:Mo-Cap

[exp(pi*sqrt(163))]

My roommate is a digital animator and if his comments are worth anything then Mo-Cap is not all it's cracked up to be

Here's how it works. You employ animators to animate humans, or you use mocap. Suppose mocap produces better results. How likely are the animators to admit it?

The truth is that good results often require a blend of human animation and mocap, But dealing with mocap is more technical than just hand animating. So for most artists mocap is hard to do well, and less interesting. So artists bitch and moan about mocap despite the fact that they can't even come close to photorealistic results without it.