Center for Human Modeling and Simulation
Document Type
Journal Article
Date of this Version
January 2000
Abstract
We describe a Parameterized Action Representation (PAR) designed to bridge the gap between natural language instructions and the virtual agents who are to carry them out. The PAR is therefore constructed based jointly on implemented motion capabilities of virtual human figures and linguistic requirements for instruction interpretation. We will illustrate PAR and a real-time execution architecture controlling 3D animated virtual human avatars.
Download
Date Posted: 13 July 2007
Comments
Postprint version. Published in American Association for Artificial Intelligence, Spring Synposium, 2000.