Surface Reconstruction Preserving Discontinuities.

Abstract

This paper presents some experimental results that indicate the plausibility of using non-convex variational principles to reconstruct piecewise smooth surfaces from sparse and noisy data. This method uses prior generic knowledge about the geometry of the discontinuities to prevent the blurring of the boundaries between continuous subregions. The author includes examples of the application of this approach to the reconstruction of dynthetic surfaces, and to the interpolation of disparity data from the stereo processing of real images. (Author)

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 01, 1984
Accession Number
ADA146741

Entities

People

  • J. L. Marroquin

Organizations

  • Massachusetts Institute of Technology

Tags

Communities of Interest

  • Energy and Power Technologies
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Boundaries
  • Computer Graphics
  • Computer Stereo Vision
  • Computer Vision
  • Detection
  • Detectors
  • Discontinuities
  • Disparities
  • Geometry
  • Image Processing
  • Information Processing
  • Interpolation
  • Military Research
  • Probability
  • Stochastic Processes
  • Variational Principles

Readers

  • Computer Vision.
  • Finite Element Method (FEM) for solving Partial Differential Equations (PDEs)