Motionscan is a new CG production process where an actor’s performance is captured 1:1 without any post capture interpretation. Literally what you see on screen is what was captured from the actors performance. Instead of tracking balls or other types of markers MotionScan captures the actual face thereby revealing to your audience the humanity of a performance with all its foibles, frailty, strength and nuance. The finished product appears compellingly lifelike as opposed to current technologies where the eyes, mouth, neck and forehead all look independently animated. Motionscan was used to capture over 400 actors for Rockstar and Team Bondi’s incredibly successful console game L.A.Noire.
MotionScan is also incredibly fast to use. Capture is real time and data processing is a function of the number of computers available in the processing pipeline not how many animators are available and can run 24 hours a day. L.A Noire featured 21 hours of captured MotionScan data. Don’t like the take you just captured? Captured some more. It takes as long as takes for the the actor to say their line. Capture lots of variations so that you have more choices. Remember it’s real time and not limited by available labour. Motionscan data is also scalable for any type of device and Level of Detail can be modified in real time during decompression.
If you want real people to give real performances that resonate with your audience in your next video game, CG movie, pre viz or trailer MotionScan could be for you.
This clip is an excerpt from “The Technology Behind Performance” which is copyright Rockstar Games And Take Two Interactive Software Inc.