Metamorphic image registration using a semi-Lagrangian scheme

DOI10.1007/978-3-030-80209-7_84zbMATH Open1485.94005arXiv2106.08817OpenAlexW3170098260MaRDI QIDQ2117947FDOQ2117947

Authors: A. François, Pietro Gori, Joan Alexis Glaunès 

Publication date: 22 March 2022

Abstract: In this paper, we propose an implementation of both Large Deformation Diffeomorphic Metric Mapping (LDDMM) and Metamorphosis image registration using a semi-Lagrangian scheme for geodesic shooting. We propose to solve both problems as an inexact matching providing a single and unifying cost function. We demonstrate that for image registration the use of a semi-Lagrangian scheme is more stable than a standard Eulerian scheme. Our GPU implementation is based on PyTorch, which greatly simplifies and accelerates the computations thanks to its powerful automatic differentiation engine. It will be freely available at https://github.com/antonfrancois/Demeter_metamorphosis.

Full work available at URL: https://arxiv.org/abs/2106.08817

Recommendations

• A Lagrangian Gauss-Newton-Krylov solver for mass- and intensity-preserving diffeomorphic image registration
• Diffeomorphic 3D image registration via geodesic shooting using an efficient adjoint calculation
• Spatially adaptive metrics for diffeomorphic image matching in LDDMM
• Diffeomorphic image matching with left-invariant metrics
• Fast diffeomorphic image registration via Fourier-approximated Lie algebras

zbMATH Keywords

semi-Lagrangian schememetamorphosisLDDMMimage diffeomorphic registration

Mathematics Subject Classification ID

Image processing (compression, reconstruction, etc.) in information and communication theory (94A08) Computing methodologies for image processing (68U10)

Cites Work

• Computing large deformation metric mappings via geodesic flows of diffeomorphisms
• Shapes and diffeomorphisms
• Metamorphosis of images in reproducing kernel Hilbert spaces
• Variational problems on flows of diffeomorphisms for image matching
• The Euler-Poincaré theory of metamorphosis
• Local Geometry of Deformable Templates
• Geodesic shooting for computational anatomy
• Diffeomorphic 3D image registration via geodesic shooting using an efficient adjoint calculation
• A computational realization of a semi-Lagrangian method for solving the advection equation
• Image reconstruction through metamorphosis

Cited In (2)

• An operator-splitting approach for variational optimal control formulations for diffeomorphic shape matching
• HV geometry for signal comparison

Uses Software

• PyTorch
• GitHub

This page was built for publication: Metamorphic image registration using a semi-Lagrangian scheme

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2117947)