PhD Program Summer School 2006
Topics on scientific visualization
| speaker | Prof. Dr. Ken Joy, University of California, Davis, Cal., USA |
1. Visualization of Material Boundaries
| date | September 27, 2006 |
| abstract | In this talk, we present a review of algorithms for material interface reconstruction on data sets where material fractions are given for each cell or data point in the grid. We will present several methods by which reconstruction problem is transformed to a problem similar to isosurface reconstruction. Methods of triangulating the material boundary surface are discussed, along with methods to minimize the error. Applications of the method will be discussed, as well as future research problems. |
2. Interactive Rendering of Planetary-Scale Geometry and Texture
| date | September 28, 2006 |
| abstract | The real-time display of huge geometry and imagery databases involves view-dependent approximations, typically through the use of precomputed hierarchies that are selectively refined at runtime. This talk focuses on the problem of terrain visualization, in which planetary databases involving billions of elevation and color values are displayed in PC graphics hardware at high frame rates. We show how innovative data structures, new out-of-core storage organization based on space-filling curves, and optimization using graphics processors can be used to solve this problem. New aerial surveillance methods produce massive video streams that must be displayed in real time. We discuss methods by which these massive video streams can be visualized. |
3. Thinking Outside the Visualization Box
| date | September 29, 2006 |
| abstract | Are we working on the correct problems? Are we addressing the new fundamental challenges that allow scientists and engineers to explore massive time-varying multi-valued data sets? Or are we pushing incremental updates to prior solutions? Currently, the field of visualization is largely driven by improvements in technology, and not by new visualization solutions that are necessary to makes fundamental advances in discovery and analysis of complex data. I believe that we need to think "outside the visualization box," identifying new "visualization" research challenges, not continue to provide incremental updates to existing solutions. In this talk, we identify several areas where "Thinking outside the visualization box" may significantly help. |