VDSlib: view-dependent polygonal simplification package

David Luebke luebke at cs.virginia.edu
Fri Feb 4 15:54:35 PST 2000

[I recently posted the below announcement on comp.graphics.algorithms; 
 a couple of people have since recommended that I post it to this list.]
I'd like to announce the alpha release of VDSlib, a public-domain
package for view-dependent simplification and rendering of polygonal


Geometric simplification, also known as "level of detail" or
"multiresolution modeling", is a well-known technique for managing
scene complexity in interactive rendering.  The most common approach
creates several discrete levels of detail (LODs) for every object in a
preprocess, and selects at run-time which LOD will represent each
object based on factors such as the object's distance or size on the
screen.  View-dependent simplification (VDS) takes a different
approach, creating a data structure which can be queried at run time to
generate a simplified object spanning multiple levels of detail.  Since
the simplification is adjusted dynamically, it can exploit the current
view parameters.  For example, VDS can represent portions of a large
object near the viewer with high fidelity but distant portions at low
fidelity, or or render silhouette regions of an object at more detail
than interior regions.

VDSlib provides a software library, written in C, for the construction 
and rendering of view-dependent simplifications.  It is designed to be
extremely flexible, with a callback-based API that allows developers or
researchers to plug in their own criteria for simplifying, culling, and
rendering the model.

VDSlib is based on the algorithm presented in the SIGGRAPH 97 paper
"View-Dependent Simplification of Arbitrary Polygonal Environments" by
David Luebke and Carl Erikson, with a number of improvements and
refinements.  This algorithm is similar in some ways to other
view-dependent simplification schemes introduced independently by
researches such as Hughes Hoppe (Progressive Meshes) and Xia & Varshney
(Merge Trees).  We hope that VDSlib, with its flexible framework, will
prove useful both to 3-D developers wishing to experiment with the
latest LOD techniques and to researchers interested in exploring
algorithms, criteria, and metrics for view-dependent simplification.

For more details, including full documentation and a sample application, 
please see the VDSlib web site at http://vdslib.virginia.edu.


Dave Luebke
David Luebke, Assistant Professor                University of Virginia
http://www.cs.virginia.edu/~luebke               Department of Computer Science
luebke at cs.virginia.edu                           Charlottesville, VA 22903-2442

The compgeom mailing lists: see
or send mail to compgeom-request at research.bell-labs.com with the line:
send readme
Now archived at http://uiuc.edu/~sariel/CG/compgeom/maillist.html.

More information about the Compgeom-announce mailing list