Distance functions on noisy point clouds

Guillermo Sapiro
University of Minnesota
Mathematics

During this talk, I will present theoretical and
computational results for finding distance functions and
geodesics on hyper-surfaces defined by noisy samples (point clouds).
We will show how this can be done without intermediate
surface reconstruction, in a computationally optimal fashion.
This is one of the most basic operation needed to
work with this very important source of data.
Applications range from graphics to biology to finance.

Joint work with PhD student Facundo Memoli.


Back to Multiscale Geometric Analysis: Theory, Tools, and Applications