 
Summary: GPUassisted Surface Reconstruction on
Locallyuniform Samples
Yong Joo Kil and Nina Amenta
University of California, Davis.
kil@cs.ucdavis.edu
amenta@cs.ucdavis.edu
Summary. In pointbased graphics, surfaces are represented by point clouds with
out explicit connectivity. If the distribution of the points can be carefully controlled,
surface reconstruction becomes a much easier problem. We present a simple, com
pletely local surface reconstruction algorithm for input point distributions that are
locally uniform. The locality of the computation lets us handle large point sets using
parallel and outofcore methods. The algorithm can be implemented robustly with
floatingpoint arithmetic. We demonstrate the simplicity, efficiency, and numerical
stability of our algorithm with an outofcore and parallel implementation using
graphics hardware.
1 Introduction
The idea of pointbased graphics is that a point sample can be the primary
representation of a surface, simplifying computation and saving space by do
ing without explicit connectivity information. In some situations, for instance
a geometric modeling system or the simulation of a moving front, the distri
