Reducing the memory required to find a geodesic shortest path on a large mesh

  • Authors:

  • Vishal Verma;Jack Snoeyink

  • Affiliations:

  • University of North Carolina at Chapel Hill;University of North Carolina at Chapel Hill

  • Venue:
  • Proceedings of the 17th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

Finding shortest paths and distances on the surface of a mesh in R3 is a well studied problem, with most research aiming to minimize computation time. However for large meshes, such as TIN terrain models in GIS, the major bottleneck is often the memory required by an algorithm. In this paper, we evaluate techniques for computing path distances (both for paths restricted to edges of the mesh and for paths traveling freely across the triangles of the mesh) that do not need to store data structure for the entire mesh in memory. In particular, we implement a novel combination of Dijkstra, A*, and MMP (aka, continuous Dijkstra) methods that, in our experiments on TINs containing millions of triangles, reduces the memory requirement by two orders of magnitude. We are also able to compare distances computed by Dijkstra, fast marching, and the MMP methods.