Holy shit, are we finally about to leave the uncanny valley? (LA Noir)

[Gregness]

I'm normally not one to get overly excited about new graphics developments, but These are some legitimately believeable facial expressions (http://www.youtube.com/watch?v=iiHecVjYtBY).

[EVILNess]

Oh my!

I seem to have the vapors!

[Aerozord]

stuff like this is why it bothers me that people still want higher graphic fidelity. Let me clarify, not that this isn't definite step up, but that we already have our tools, we just have to learn how to use them better

[Kyanbu The Legend]

And Bondie ups the graphic scale with some awesome visuals. we're getting rather close to the peak of CG.

[stefan]

skin's still a bit fucked up, but the expressions are down perfectly.

[Shyria Dracnoir]

Skin might be more due to the limits of the lighting in the game itself. Looking at the side by side comparisons they did at 1:50, it's very, very well done.

This is a bit of a tangent, but is anyone else wondering how this technology could be applied in other areas? I could see it being used to treat people with Aspergers; a few big traits of people on the autistic spectrum are difficulty recognizing things like nonverbal language and facial expressions. Interacting with a system like this would help them improve their skills, especially if they're leery of doing it with another person (although that would be the goal in the end).

[Loyal]

Wow. That's some seriously kickass animation.

[Aerozord]

This is a bit of a tangent, but is anyone else wondering how this technology could be applied in other areas? I could see it being used to treat people with Aspergers; a few big traits of people on the autistic spectrum are difficulty recognizing things like nonverbal language and facial expressions. Interacting with a system like this would help them improve their skills, especially if they're leery of doing it with another person (although that would be the goal in the end).

well would help with practice, but ultimately it would present the same issues. Aspies prefer digital communication specifically because it forces everyone onto the same playing field. Not an inherent hatred of communicating with others, in fact its normally a desire to socialize and internet just makes it possible.

[Doc ock rokc]

I bet you could get similar results with a few Kinect's to have a depth mask animation model.

[Azisien]

The closer we get to realism, the more we can pick it apart.

The facial animations in this game are incredible. But there's still plenty of flaws. A lot of that comes from running on 2005/2006 technology too. Zero or barely any AA/filtering. Slap this on a good rig of hardware now and it'll purr.

[Aerozord]

I bet you could get similar results with a few Kinect's to have a depth mask animation model.

my understanding is Kinect isn't sensitive enough to even pick up obvious facial features like the nose, let alone cheek and lip movement. Though they will probably increase it over time

[Fifthfiend]

Nope.

[bluestarultor]

my understanding is Kinect isn't sensitive enough to even pick up obvious facial features like the nose, let alone cheek and lip movement. Though they will probably increase it over time

Maybe with newer versions, but I really doubt it's happening with the current model.

Also, to be frank, I think it looks like crap. The animation is fine, but the faces themselves look less real and more like terrible rubber masks. So essentially a hobo with a Rolex.

That's not to say the technology isn't amazing, but the graphics the tech is being used for could be much, much better. Then again I never did like how the Xbox line handled diffuse lighting for some reason.

[Bells]

yeah let me play devil's advocate here for a second.

First of all... L.A. Noir is the hi-def versio of this (http://www.youtube.com/watch?v=pAQ-C1suCEY) (Head's up that's Phantasmagoria Footage)

Now, it's not fair to shit on this. One reason is because this is the first time this tech is used. And the first time any tech is used there is always flaws by the ton. And this doesn't even have that many flaws.

But what i see is a little bit of Floating Head syndrome and a 3d plaster of a Actor's face. Thus Hi-def Phantasmagoria. One of the good things of motion capture is that the Actor can be anyone. Let's go back to Mass Effect... which has some DREADFUL looking faces... yet, Aliens look awesome. With this tech, right now, they couldn't do that game because the actor is showed like he is in the game, they don't show any sort of chance of visual manipulation of the character.

It's really What-you-see-is-what-you-get.

So... even though it looks fantastic, it's more limited than stuff we already do and have. So no real Paradigm shift yet....

The floating head thing is most likely because they record every scene twice (acting and then faces) which again works as a point (a Arguable point...) to this tech not being much more effective than what we already have...

Also i would like to point another technique that gathered great results... the Movie RANGO a fully animated, non-pixar movie. Which was hailed all around as being incredible in the animation department... how was that movie done?

Like this (http://www.youtube.com/watch?v=r-Bc43WVoL0)

So, still a more traditional take and it gives great results... all you really need is good use of the assets we already have. This tech can evolve to great stuff, but ain't great stuff yet