A Scale-Space Approach to Landmark Constrained Image Registration

Authors:
Eldad Haber;Stefan Heldmann;Jan Modersitzki
Affiliations:
Dept. of Math. and Computer Science, Emory Emory University, Atlanta, USA;Inst. of Mathematics, University of Lübeck, Lübeck, Germany;Dept. of Computing and Software, McMaster University, Hamilton, Canada
Venue:
SSVM '09 Proceedings of the Second International Conference on Scale Space and Variational Methods in Computer Vision
Year:
2009

Citing 9
Cited 2

Principal Warps: Thin-Plate Splines and the Decomposition of Deformations

IEEE Transactions on Pattern Analysis and Machine Intelligence
Matrix computations (3rd ed.)

Matrix computations (3rd ed.)
Multigrid

Multigrid
A Theoretical Framework for Convex Regularizers in PDE-Based Computation of Image Motion

International Journal of Computer Vision
Alignment by maximization of mutual information

Alignment by maximization of mutual information
A Multilevel Method for Image Registration

SIAM Journal on Scientific Computing
A Lattice-Preserving Multigrid Method for Solving the Inhomogeneous Poisson Equations Used in Image Analysis

ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part II
Motion compensated video super resolution

SSVM'07 Proceedings of the 1st international conference on Scale space and variational methods in computer vision
Brain image registration using cortically constrained harmonic mappings

IPMI'07 Proceedings of the 20th international conference on Information processing in medical imaging

Using landmarks as a deformation prior for hybrid image registration

DAGM'11 Proceedings of the 33rd international conference on Pattern recognition
A landmark-based primal-dual approach for discontinuity preserving registration

MICCAI'12 Proceedings of the 4th international conference on Abdominal Imaging: computational and clinical applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Adding external knowledge improves the results for ill-posed problems. In this paper we present a new multi-level optimization framework for image registration when adding landmark constraints on the transformation. Previous approaches are based on a fixed discretization and lack of allowing for continuous landmark positions that are not on grid points. Our novel approach overcomes these problems such that we can apply multi-level methods which have been proven being crucial to avoid local minima in the course of optimization. Furthermore, for our numerical method we are able to use constraint elimination such that we trace back the landmark constrained problem to a unconstrained optimization leading to an efficient algorithm.