Share Email Print

Proceedings Paper

Structural Stereo Matching Of Laplacian-Of-Gaussian Contour Segments For 3D Perception
Author(s): K. L. Boyer; G. E. Sotak
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

We solve the stereo correspondence problem using Lapla-cian of Gaussian (LoG) zero-crossing contours as a source of primitives for structural stereopsis, as opposed to traditional point-based algorithms. For each image in the stereo pair, we apply the LoG operator, extract and link zero crossing points, filter and segment the contours into meaningful primitives, and compute a parametric structural description over the resulting primitive set. We then apply a variant of the inexact structural matching technique of Boyer and Kak Ill to recover the optimal interprimitive mapping (correspon-dence) function. Since an extended image feature conveys more information than a single point, its spatial and photometric behavior may be exploited to advantage; there are also fewer features to match, resulting in a smaller combinatorial problem. The structural approach allows greater use of spatial relational constraints, which allows us to eliminate (or reduce) the coarse-to-fine tracking of most point-based algorithms. Solving the correspondence problem at this level requires only an approximate probabilistic characterization of the image-to-image structural distortion, and does not require detailed knowledge of the epipolar geometry.

Paper Details

Date Published: 7 March 1989
PDF: 8 pages
Proc. SPIE 1005, Optics, Illumination, and Image Sensing for Machine Vision III, (7 March 1989); doi: 10.1117/12.949048
Show Author Affiliations
K. L. Boyer, The Ohio State University (United States)
G. E. Sotak, The Ohio State University (United States)

Published in SPIE Proceedings Vol. 1005:
Optics, Illumination, and Image Sensing for Machine Vision III
Donald J. Svetkoff, Editor(s)

© SPIE. Terms of Use
Back to Top