Share Email Print
cover

Proceedings Paper

Modeling aspects of the surface reconstruction problem
Author(s): Charles K. Toth; Gabor Melykuti
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The ultimate goal of digital photogrammetry is to automatically produce digital maps which may in turn form the basis of GIS. Virtually all work in surface reconstruction deals with various kinds of approximations and constraints that are applied. In this paper we extend these concepts in various ways. For one, matching is performed in object space. Thus, matching and densification (modeling) is performed in the same reference system. Another extension concerns the solution of the second sub-problem. Rather than simply densifying (interpolating) the surface, we propose to model it. This combined top-down and bottom-up approach is performed in scale space, whereby the model is refined until compatibility between the data and expectations is reached. The paper focuses on the modeling aspects of the surface reconstruction problem. Obviously, the top-down and bottom-up model descriptions ought to be in a form which allows the generation and verification of hypotheses. Another crucial question is the degree of a priori scene knowledge necessary to constrain the solution space.

Paper Details

Date Published: 17 August 1994
PDF: 6 pages
Proc. SPIE 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision, (17 August 1994); doi: 10.1117/12.182852
Show Author Affiliations
Charles K. Toth, Ohio State Univ. (United States)
Gabor Melykuti, Technical Univ. of Budapest (Hungary)


Published in SPIE Proceedings Vol. 2357:
ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision
Heinrich Ebner; Christian Heipke; Konrad Eder, Editor(s)

© SPIE. Terms of Use
Back to Top