Here is some perspective: regular motion capture tends to use roughly 90 points of reference. In layman’s terms, there are 90 of those weird white dots stuck to the participants face. Camilla Luddington, who now plays Lara Croft just tweeted the above pic. Crystal Dynamics is now using a brand new technique for motion capture. That technique is fluorescent paint; paint which provides over 7000 points of reference.
#tombraider fans.This is #mova.Florescent paint that is sprayed all over my face 2give 7000 points of ref!!! pic.twitter.com/rYuiy9m6zX
— Camilla Luddington (@camilluddington) August 7, 2014
Pretty insane. According to Camilla, the paint takes under an hour to apply and, amazingly, the paint can only be seen with black light. It’s invisible to the naked eye.
Sounds absolutely incredible. I wonder just how much it will add to capture in video games? Hopefully it will herald a more complete ability for actors to translate their performance more accurately. When games have done it correctly — Uncharted, Heavy Rain, LA Noire — it really added a new dimension to certain aspects of gaming. I’d love to see a game like Tomb Raider push those boundaries.
Comments
28 responses to “Tomb Raider Now Uses Weird Paint For Next-Gen Performance Capture”
So weird seeing her on Grey’s Anatomy after playing Tomb Raider.
Not that I watch Grey’s Anatomy.
*cough*
Even weirder if you saw all of her on Californication first…
Oh she was just awful on Grey’s…
Is that show still running?
Isn’t this like the thing they mount on store doors to tag and catch shoplifters?
I’ve found a level of self control deep inside me that I never knew existed until I stopped myself from making the obvious joke about this photo.
Yeah, I know what you mean.
You wouldn’t want to be up to any shenanigans before turning up for work that day.
“That’s nothing, you should get a black-light in here. Make the place look like a Jackson Pollock painting”
Ahhh, Groot.
Hahahahaha!
I had to try very hard not to draw obvious CSI comparisons to the phrase “…can only be seen with black light…” and failed.
Must… not… ungh
I’m not saying that’s what it is or anything, but if they used porn stars instead they could just call them in after work and save themselves the cost of the makeup/spray artist.
Is this all really necessary?
If it brings the level of facial capture closer to the level of LA Noire, then why not?
But how close will you be looking in an action game?
I don’t think that really matters. It might seem like loads of effort but really, does this not seem so much easier than someone animating facial animations by hand or by some shitty automated process?
They have to do it either way, whether it’s an action game or another LA Noire. They might as well do it properly because it’s probably roughly the same amount of effort either way.
It’s probably less effort than putting 90 dots on her face… a couple of quick sprays and you’re away. Taking it off will be easier too.
So will this make games cost more? More reference points will eventually mean a game will need better looking actors/actresses to play main roles.
Games already employ models and shiz to provide character models/faces for them. The face of default BroShep in Mass Effect is a male model.
Yeah, because without revolutionary facial captures all our game characters have been voiced by a bunch of total uggos.
I agree. That last uggo in particular…
Nah, depends how the models are rigged in the software.
In a more serious frame of mind, I remember seeing this at Siggraph, I think back in 2007, so it is a fairly established technology. It was used on Benjamin Button and other films.
*edit* found a link. http://www.nytimes.com/2006/07/31/technology/31motion.html?_r=0
http://grandtextauto.org/2006/07/31/contour-a-novel-technique-for-modeling-and-capture/
This isn’t exactly new, the speckled paint option’s been around for a while. Nice to see that more companies are using it over the old face marker option where they just put dots at regular intervals on the face. There’s also a markerless option which pinpoints your nose, mouth, eyes, eye brows and a few other areas and then just maps it to the character model … but it’s kind of hit and miss at the moment
Edit: Reply fail.
Is it just me or does that picture look a bit like a raver version of Girl with a Pearl Earring?
Nah, I see it: the pose, the head band. Just crazy UV flecks instead oaf an earring…