Orthonormal Kernel Kronecker Product Graph Mdatching

  • Barend Jacobus van Wyk
  • Michaël Antonie van Wyk
Conference paper

DOI: 10.1007/3-540-45028-9_10

Part of the Lecture Notes in Computer Science book series (LNCS, volume 2726)
Cite this paper as:
Wyk B.J.., Wyk M.A.. (2003) Orthonormal Kernel Kronecker Product Graph Mdatching. In: Hancock E., Vento M. (eds) Graph Based Representations in Pattern Recognition. GbRPR 2003. Lecture Notes in Computer Science, vol 2726. Springer, Berlin, Heidelberg

Abstract

This paper introduces a novel algorithm for performing Attributed Graph Matching (AGM). The Orthonormal Kernel Kronecker Product Graph Matching (OKKPGM) algorithm is based on the recently introduced Kronecker Product Graph Matching (KPGM) formulation. However, unlike previous algorithms based on the KPGM formulation which avoided explicit compatibility calculations, the OKKPGM algorithm obtains an estimate to the Kronecker Match Matrix (KMM) using kernel function evaluations. A permutation sub-matrix (match matrix), is then inferred from the KMM using a summation procedure.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Barend Jacobus van Wyk
    • 1
    • 2
  • Michaël Antonie van Wyk
    • 2
  1. 1.Kentron, a Division of DenelCenturionSouth Africa
  2. 2.Rand Afrikaans UniversityJohannesburgSouth Africa

Personalised recommendations