Generating Virtual Avatars with Personalized Walking Gaits using Commodity Hardware.

We present a novel algorithm for generating virtual avatars which move like the represented human subject, using inexpensive sensors & commodity hardware. Our algorithm is based on a perceptual study that evaluates self-recognition and similarity of gaits rendered on virtual avatars. We identify discriminatory features of human gait and propose a data-driven synthesis algorithm that can generate a set of similar gaits from a single walker. These features are combined to automatically synthesize personalized gaits for a human user from noisy motion capture data. The overall approach is robust and can generate new gaits with little or no artistic intervention using commodity sensors in a simple laboratory setting. We demonstrate our approach's application in rapidly animating virtual avatars of new users with personalized gaits, as well as procedurally generating distinct but similar "families" of gait in virtual environments.