Visually realistic simulation of physical systems has become an important part of many computer graphics applications, and there have been tremendous improvements in quality over the last decade. Despite these improvements, there are several remaining challenges. For one, there remains a very large gap between offline simulation and real-time simulation. We would like to have systems that can respond to user input in a physically-believable fashion, at a rate sufficient to allow a user to interact with the system as it develops. Another remaining challenge is in providing user control to simulations. For graphics applications, our goal when simulating is often to create a plausible desired outcome, rather than a pure, physically-precise result.
This talk will discuss some of the recent techniques developed in our group's ongoing work in interactive and controllable simulation for graphics applications. By modeling and simulating specific higher-level effects rather than simulating strictly from first principles, we can achieve faster performance with minimal sacrifices in visual quality. Among the techniques we have used are those based on statistically-based simulation, use of proxy objects, level-of-detail simulation, and particle-based effects. Application of these methods to a variety of problems will be discussed, focusing mainly on simulations of large numbers of rigid bodies, water behavior, burning objects, and plant motion. We will also discuss some recent work on controlling and guiding simulation, in the context of fluid simulation and rigid body simulation.
John Keyser is an Associate Professor and the Associate Department Head for Academics in the Department of Computer Science at Texas A&M University. He received his Ph.D. in Computer Science from the University of North Carolina (Chapel Hill) in 2000, and B.S. degrees in Engineering Physics, Applied Mathematics, and Computer Science from Abilene Christian University in 1994. His research has touched on a wide range of graphics topics, ranging from early work on robust geometric calculations, to more recent work related to rendering. In addition to his current project in physically based simulation, he is involved with a project focused on visualization and reconstruction of geometric data from very large scanned data sets. He serves as a member of the Executive Committee of the Solid Modeling Association, and as an Associate Editor for IEEE Transactions on Visualization and Computer Graphics.