WELCOME TO THE ON-LINE HOME OF TEDxNEWYORK.
TEDxNewYork is held at GreyNY, 200 Fifth Avenue. We meet every week (mostly) on Fridays now and (mostly) from 1-2pm. We are open to the public. If you want to attend, send a note to firstname.lastname@example.org (that's Don McKinney & Chel O'Reilly) with your vitals. Our biggest limitation is space so give us plenty of notice and we'll do our best to accommodate. Hope to see you at one of our events soon.
Thursday, February 5, 2009
Session 4: See- Ed Ulbrich
Showed us a film clip from Benjamin Button... Ed helped computer generate Brad Pitt for the FIRST HOUR of the movie (couldn't tell, could ya!)
This project wasn't possible 5 years ago, and animating the human head is the holy grail of rendering.
The vision was to have the same actor play the character throughout the film entirely.
This was a high risk project. The process of creation was a 12 step program
1: admit you have a problem
2: break the problem down (3 areas- age brad pitt (and his little ticks, etc; give him all emotions; place him anywhere in time/space). motion tracking is used to capture is good, but they need far more than what it could give
Facial Action Coding System- a way of seeing 70 different facial poses that can be combined in so many ways, we can create the world of a person's emotions. They filmed Brad at all 70 expressions.
Contour camera process of painting face in phosphorescence to gather a kazillion points of data instead of the grid that motion capture points.
He's now unvieling the physical life casts of Brad aged in a variety of ways. Then THESE were scanned for additional data.
Now they needed to combine Brad's performance and match it to the digital information they'd gathered to make the scene.
I wasn't interested in the movie but now I HAAAAVE to see it.
Oh, and for their data gathering they needed to gather lighting info.
The eyes were most important - someone was devoted to Brad's eyes for 2 years then another to his tongue for 9 months (yes we all laughed like junior high school students).
They called this "emotion capture". And it is beautiful. And it only took 150 people 2 years.