Broad Area Colloquium For AI-Geometry-Graphics-Robotics-Vision
(CS 528)

Geometric Optimization with Moving and Deformable Objects

Helmut Pottmann, Technische Universitaet Wien
November 22, 2004, 4:15PM
TCSeq 200


In Geometric Computing there is a variety of application areas for geometric optimization algorithms that deal with the positioning or deformation of geometric objects. Examples include data fitting in Computer Aided Geometric Design, segmentation and registration problems in Image Processing and Computer Vision, and positioning problems in Robotics. The speaker will present recent progress in the solution of these optimization problems. One important ingredient is a careful geometric investigation of the distance function of a geometric object. Another key aspect is an investigation of the convergence behavior of standard algorithms for the addressed problems and the development of new algorithms with faster convergence and improved stability. The presented results are based on joint work with M. Hofer and S. Leopoldseder (Vienna), Q. Huang, Y. Yang and S. Hu (Tsinghua Univ., Beijing), W. Wang and Y. Liu (Univ. of Hong Kong) and N. Mitra, N. Gelfand and L. Guibas (Stanford).

About the Speaker

Helmut Pottmann studied mathematics and descriptive geometry at Vienna University of Technology, and received the PhD in 1983, the habilitation in geometry in 1986. Between 1988 and 1992 he held faculty positions at University of Kaiserslautern, Purdue University and University of Hamburg. Since 1992, he is professor of geometry at Vienna University of Technology. His research interests include classical geometry and its applications, especially to Computer Aided Geometric Design, Kinematics, Computer Vision, and the relations between Geometry, Numerical Analysis and Approximation Theory. He is associate editor of the journals Computer Aided Design, Computer Aided Geometric Design and The Visual Computer.


Back to the Colloquium Page