Share Email Print

Proceedings Paper

Stereo Matching As Constrained Optimization Using Scale Continuation Methods
Author(s): Demetri Terzopoulos; Andrew Witkin; Michael Kass
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

A new stereo matching algorithm is developed and applied to natural scenes. The algorithm is based on a recent approach to matching multiple, multidimensional signals that have been deformed with respect to one another.' The goal is to optimally recover the deformation map, which in this case represents the stereoscopic disparity between the left and right images. The problem is formulated as the minimization of an energy measure that combines a similarity functional with a controlled-continuity constraint. Applying the continuation method, this nonlinear, nonconvex minimization problem is solved by a deterministic dynamic system governed by a set of coupled, first-order differential equations. The system finds an optimal approximation at a coarse scale, then tracks it continuously to a fine scale, thus avoiding bad local minima. The stereo algorithm succinctly unifies the notions of matching as constrained optimization, of coarse-to-fine search, and of variational surface reconstruction.

Paper Details

Date Published: 21 August 1987
PDF: 8 pages
Proc. SPIE 0754, Optical and Digital Pattern Recognition, (21 August 1987); doi: 10.1117/12.939972
Show Author Affiliations
Demetri Terzopoulos, Schlumberger Palo Alto Research (United States)
Andrew Witkin, Schlumberger Palo Alto Research (United States)
Michael Kass, Schlumberger Palo Alto Research (United States)

Published in SPIE Proceedings Vol. 0754:
Optical and Digital Pattern Recognition
Hua-Kuang Liu; Paul S. Schenker, Editor(s)

© SPIE. Terms of Use
Back to Top