A method for the transport and registration of images on implicit surfaces

Authors:
Christophe Chefd'hotel
Affiliations:
Siemens Corporate Research, Princeton, NJ
Venue:
SSVM'07 Proceedings of the 1st international conference on Scale space and variational methods in computer vision
Year:
2007

Citing 6
Cited 0

Geodesic Active Contours

International Journal of Computer Vision
Diffeomorphisms Groups and Pattern Matching in Image Analysis

International Journal of Computer Vision
Variational problems and partial differential equations on implicit surfaces

Journal of Computational Physics
A Level-Set Based Approach to Image Registration

MMBIA '00 Proceedings of the IEEE Workshop on Mathematical Methods in Biomedical Image Analysis
Landmark Matching via Large Deformation Diffeomorphisms on the Sphere

Journal of Mathematical Imaging and Vision
Deformable templates using large deformation kinematics

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Image transport is at the heart of large deformation registration algorithms and level set methods. Here we propose to extend this concept to solve alignment and correspondence problems for images defined on general geometries. Our approach builds upon the framework proposed by Bertalmio et al. [1] for solving PDEs on implicit surfaces. The registration process is defined by a system of transport equations and is driven by the gradient of a similarity functional. The transformation is regularized using a nonlinear heat equation. Compared to recent developments for image registration on manifolds [4], the implicit representation of the image domain allows us to deal easily with arbitrary surfaces. We illustrate the potential of this technique with several synthetic experiments.