General Object Reconstruction Based on Simplex Meshes

  • Authors:
  • Hervé Delingette

  • Affiliations:
  • INRIA, B.P. 93, 2004 route des Lucioles, 06902 Sophia Antipolis Cedex, France. Herve.Delingette@inria.fr

  • Venue:
  • International Journal of Computer Vision
  • Year:
  • 1999

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we propose a general tridimensionalreconstruction algorithm of range and volumetric images, based ondeformable simplex meshes. Simplex meshes are topologically dual oftriangulations and have the advantage of permitting smoothdeformations in a simple and efficient manner. Our reconstructionalgorithm can handle surfaces without any restriction on their shapeor topology. The different tasks performed during the reconstructioninclude the segmentation of given objects in the scene, theextrapolation of missing data, and the control of smoothness,density, and geometric quality of the reconstructed meshes. Thereconstruction takes place in two stages. First, the initializationstage creates a simplex mesh in the vicinity of the data model eithermanually or using an automatic procedure. Then, after a fewiterations, the mesh topology can be modified by creating holes or byincreasing its genus. Finally, an iterative refinement algorithmdecreases the distance of the mesh from the data while preservinghigh geometric and topological quality. Several reconstructionexamples are provided with quantitative and qualitative results.