I've moved

That's right, my (successful) thesis defense took place on December 14, 1998 and I produced the final copy of my dissertation on May 9, 1999. I left CMU at the end of July 1999. I'm now an assistant professor in the Department of Computer Science at the University of Illinois at Urbana-Champaign. My current home page can be found at

My Web pages here will continue to be available for a while yet. But I won't be updating them after I've migrated their content to my new location.

The Home of Michael Garland

Find out why Beethoven is wearing a blue ribbon.

      From September 1993 until I graduated, I was a graduate student in the CMU School of Computer Science. My advisor was Paul Heckbert. For your amusement, you can watch me morph into him.

Computer Graphics

Yes, that's what I do (hence the computer generated bust of Beethoven above). My current research is focused on multiresolution modeling.
Polygonal Cow Model: This is the sort of thing that I spend a lot of my time working with.
[INLINE: Gratuitous image of a synthetic bovine]

      This brings me to the cow sitting over there. When exploring multiresolution modeling techniques, I need models to experiment with. I have adopted this cow as my personal favorite test case. It's much more interesting than say a plane or a car. I elaborate on this idea on my Bovine Page. Try it out.

      So what is this multiresolution modeling thing? If you follow the link above, you'll come to my page on the subject. You can also try the survey paper, Multiresolution Modeling For Fast Rendering, listed below. Briefly put, polygonal models are by far the most common sort of models in computer graphics. In order to capture all the detail of the object being modeled, such models often utilize a large number of polygonal elements. This is necessary when the object is being viewed up close. However, if the same object is displayed at progressively greater distances, such detail becomes more and more unnecessary. For rapid rendering, we would like to display a model which captures the detail which is actually visible at the current distance but which does not model unnecessary details. There are systems which support this, but the successive levels of detail for the model must be generated by hand. The thrust of the research which I have been pursuing is to automatically create these levels of detail.


Since you're here, you might be interested in papers that I've written. Here's a list of them, and links to the Postscript versions:

Quadric-Based Polygonal Surface Simplification
by Michael Garland, Ph.D. thesis, Tech. Rept. CMU-CS-99-105.
Face Cluster Radiosity
by Andrew Willmott, Paul Heckbert, and Michael Garland. Eurographics Workshop on Rendering, June 1999.
Simplifying Surfaces with Color and Texture using Quadric Error Metrics,
by Michael Garland and Paul Heckbert, IEEE Visualization 98.
Surface Simplification Using Quadric Error Metrics,
by Michael Garland and Paul Heckbert, SIGGRAPH 97.
Free software is now available.
Survey of Polygonal Surface Simplification Algorithms,
by Paul Heckbert and Michael Garland. This draft appeared in the SIGGRAPH 97 course notes. The final revision will be published as a CMU-CS Technical Report.
Fast Polygonal Approximation of Terrains and Height Fields,
by Michael Garland and Paul Heckbert (Technical Report CMU-CS-95-181). A color plate is included separately. The whole paper is also available as a PDF file. For more information, and some free software, try looking here.
Multiresolution Modeling For Fast Rendering,
by Paul Heckbert and Michael Garland, Graphics Interface '94. The version published in Graphics Interface had certain errors that have been corrected in this online version.

Michael Garland

Last modified: Thu Aug 5 16:03:25 EDT 1999