Harmonic Analysis and Signal Processing Seminar
Multiscale Refinement Subdivision in Nonlinear and Geometric Settings
Thomas Yu, Rensselaer Polytechnic Institute
Wednesday, October 13, 2004, 2-3:00pm, WWH 1314
Abstract
Multiscale refinement subdivision are methods for taking
coarsely sampled data and recursively creating very finely sampled data consistent
with the coarse-scale data. They arise in intriguing ways from wavelet and
pyramid algorithms in signal and image processing, from geometric modelling
problems in computer aided design and manufacturing, from processing of surface
data gathered by 3D scanners and, more recently, from a variety of emerging
geometric representation problems which involve data taking values in nonlinear
manifolds.
Subdivision algorithms typically look quite simple, but their simplicity
is deceptive -- more often than not it is highly nontrivial to understand
their properties. The speaker will discuss a number of recent results and
some of the breakthrough results from the past 20 years or so, and discuss
many challenging analysis problems that remain.