---
title: "Stereo Matching by Hierarchical, Microcanonical Annealing"
authors:
  - "Stephen T. Barnard"
doi: ""
url: "https://ijcai.org/proceedings/1987/174"
pdf: "https://ijcai.org/proceedings/1987/0174.pdf"
venue: "Proceedings of the Tenth International Joint Conference on Artificial Intelligence"
year: 1987
pages: "832-834"
track: "Perception"
keywords:
---

# Stereo Matching by Hierarchical, Microcanonical Annealing

Stephen T. Barnard

*Proceedings of the Tenth International Joint Conference on Artificial Intelligence*, pages 832-834.

## Abstract

An improved stochastic stereo-matching algorithm is presented. It incorporates two substantial modifications to an earlier version: a new variation of simulated annealing that is faster, simpler, and more controllable than the conventional "heatbath" version, and a hierarchical, coarse-to-fine-resolution control structure. The Hamiltonian used in the original model is minimized, but far more efficiently. The basis of microcanonical annealing is the Creutz algorithm . Unlike its counterpart, the familiar Metropolis algorithm, the Creutz algorithm simulates a thermally isolated system at equilibrium . The hierarchical control structure, together with a Brownian state-transition function, tracks ground states across scale, beginning with small, coarsely coded levels. Results are shown for a 512 x 512 pair with 50 pixels of disparity.