Linear image registration through MRF optimization

  • Authors:

  • Ben Glocker;Darko Zikic;Nikos Komodakis;Nikos Paragios;Nassir Navab

  • Affiliations:

  • Computer Aided Medical Procedures, Technische Universität München, Germany and Laboratoire MAS, Ecole Centrale Paris, Chatenay-Malabry, France;Computer Aided Medical Procedures, Technische Universität München, Germany;Computer Science Department, University of Crete, Greece;Laboratoire MAS, Ecole Centrale Paris, Chatenay-Malabry, France and Equipe GALEN, INRIA Saclay, Orsay, France;Computer Aided Medical Procedures, Technische Universität München, Germany

  • Venue:
  • ISBI'09 Proceedings of the Sixth IEEE international conference on Symposium on Biomedical Imaging: From Nano to Macro
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose a Markov Random Field formulation for the linear image registration problem. Transformation parameters are represented by nodes in a fully connected graph where the edges model pairwise dependencies. Parameter estimation is then solved through iterative discrete labeling and discrete optimization while a label space refinement strategy is employed to achieve sub-millimeter accuracy. Our framework can encode any similarity measure, allows for automatic reduction of the degrees of freedom by simple changes on the MRF topology, and is robust to initialization. Promising results on real data and random studies demonstrate the potential of our approach.