Applied Math Seminar
Winter Quarter 2007
3:15 p.m.
Sloan Mathematics Corner
Building 380, Room 380-C


Friday, March 16, 2007

Charles K. Chui
Dept. of Statistics, Stanford University

From Interpolating Subdivisions to Representation of Scattered Data


Abstract:

In this lecture, I will introduce this topic by extending the cubic B-spline refinement equation to the bivariate setting, thereby arriving at some re-formulation of Loop s scheme to achieve certain degree of freedom. This allows us to extend Loop s approximation subdivision scheme to interpolating surface subdivisions, without increasing the cubic polynomial degree and decreasing the C^2 smoothness property. Subdivision templates for extraordinary vertices of arbitrary valences are derived accordingly. An example of our non-spline results to achieve one-ring templates will be given. I will also discuss how such interpolating subdivision templates can be adopted to construct minimum-supported basis functions for bivariate scattered data interpolation.