Least Committment Graph Matching by Evolutionary Optimisation

  • Authors:
  • Richard Myers;Edwin R. Hancock

  • Affiliations:
  • -;-

  • Venue:
  • ECCV '00 Proceedings of the 6th European Conference on Computer Vision-Part I
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents a method of matching ambiguous feature sets extracted from images. The method is based on Wilson and Hancock's Bayesian matching framework [1], which is extended to handle the case where the feature measurements are ambiguous. A multimodal evolutionary optimisation framework is proposed, which is capable of simultaneously producing several good alternative solutions. Unlike other multimodal genetic algorithms, the one reported here requires no extra parameters: solution yields are maximised by removing bias in the selection step, while optimisation performance is maintained by a local search step. An experimental study demonstrates the effectiveness of the new approach on synthetic and real data. The framework is in principle applicable to any multimodal optimisation problem where local search performs well.