Villanova Department of Computing Sciences

CSC Colloquium: Patrick Cozzi

Introduction to Massive Model Visualization

Time: Monday, March 09, 2009 at 04:30 PM
Location: MSC 102

Patrick Cozzi, Analytical Graphics, Inc.

Abstract:

Today's massive geometric models can have billions of polygons and gigabytes of textures. Such models are becoming common in Virtual Globes and simulation applications, which need to display detailed terrains, cities, aircrafts, etc. This talk introduces pragmatic techniques to render massive models on a typical PC. No background in computer graphics is assumed.

With advances in model acquisition and procedural modeling, massive model complexity continues to outpace the explosive growth of CPU and GPU processing power. Brute force rendering cannot achieve interactive frame rates. Even if massive models could fit into video memory, current GPUs can only process 10-200 million triangles per second. Interactive massive model rendering requires techniques that are output-sensitive: performance is a function of the number of pixels rendered, not the size of the model. Such techniques are presented in this talk, including visibility culling, level of detail, and memory management.

This talk is filled with images and videos, and is sure to entertain even those with no previous computer graphics knowledge.

Slides

Speaker Bio:

Patrick Cozzi, Analytical Graphics, Inc.

Patrick Cozzi is a software engineer on the 3D team at Analytical Graphics, Inc., where he develops the next generation 3D engine for the aerospace and defense industries. Before joining AGI in 2004, Patrick worked for Intel and IBM. He was awarded a patent for his work on storage systems at IBM. Patrick hold a B.S. in Computer Science from the Pennsylvania State University and an M.S.E in Computer and Information Science from the University of Pennsylvania. This talk is based on his thesis: "Visibility Driven Out-of-Core HLOD Rendering." Visit his team's blog at blogs.agi.com/pointbreak.

Refreshments will be served in MSC 159 after the talk.