Harnessing Parallelism for Interactive 3D Computer Graphics

Anselmo Lastra
Department of Computer Science
University of North Carolina at Chapel Hill

Interactive 3D graphics has proven very useful for spatially oriented
tasks, such as design, medical diagnosis, molecular biology, and
visual simulation. It has also now entered the mainstream of
computing, as evidenced by the number of 3D accelerator cards being
sold for PCs. In this talk, we first outline the type and quantity of
computation necessary to render images from 3D datasets and, to
motivate the work, show some examples of uses for 3D graphics. We then
examine the space of architectures for parallel-rendering accelerators
and focus on PixelFlow, a scalable high-performance parallel graphics
engine designed and built at the University of North Carolina and at
the Hewlett-Packard Chapel Hill Graphics Laboratory. We conclude by
examining a possible future direction for interactive graphics:
rendering from real-world images rather than geometric datasets.


Bio:

Anselmo Lastra is a Research Associate Professor of Computer Science
at the University of North Carolina at Chapel Hill. He serves as the
software manager for the Pixel-Planes/PixelFlow research team. The
research group is currently working on designs for image-based
rendering and completing PixelFlow, a scalable graphics computer
expected to perform more than an order of magnitude faster than their
previous machine, Pixel-Planes 5. Dr.  Lastra received Ph.D. and
M.S. degrees in Computer Science from Duke University and a
B.S.E.E. from the Georgia Institute of Technology. Prior to coming to
North Carolina, he was a project manager at Coulter Electronics,
leading the development of medical instrumentation, and was a
consultant at AT & T Bell Laboratories.