Communications of the ACM
Team Develops New Model For Animated Faces and Bodies
Disney researcher Tomas Simon
Credit: Disney Research
Disney Research and Carnegie Mellon University researchers have developed a way of modeling dynamic objects such as expressions on faces, gesticulations on bodies, and the draping of clothes.
They say the method enables them to create a much more compact, powerful, and easy-to-manage model that can produce a dynamic sequence after discarding 99 percent of the original data points. Their findings were recently presented at the SIGGRAPH 2012 conference.
Disney researcher Tomas Simon says the models created with the bilinear spatiotemporal representation make it easy to change one point in space-time--such as bringing the head of a soccer player forward to make contact with a ball--while keeping it consistent with other points in the model. He notes that a sequence that would take several hours for a computer graphic artist to process using conventional models could be completed in minutes using the new models, with comparable quality.
"The ability to interact with large dynamic sequences in data consistent ways and in real time has lots of interesting applications," says Disney researcher Iain Matthews. He says the bilinear spatiotemporal simulation is possible partly because contemporary computer systems have enough memory to process data sets that can contain millions of variables.
From Disney Research
View Full Article
Abstracts Copyright © 2012 Information Inc., Bethesda, Maryland, USA
No entries found