A geometric framework for stochastic shape analysis
From MaRDI portal
Abstract: We introduce a stochastic model of diffeomorphisms, whose action on a variety of data types descends to stochastic evolution of shapes, images and landmarks. The stochasticity is introduced in the vector field which transports the data in the Large Deformation Diffeomorphic Metric Mapping (LDDMM) framework for shape analysis and image registration. The stochasticity thereby models errors or uncertainties of the flow in following the prescribed deformation velocity. The approach is illustrated in the example of finite dimensional landmark manifolds, whose stochastic evolution is studied both via the Fokker-Planck equation and by numerical simulations. We derive two approaches for inferring parameters of the stochastic model from landmark configurations observed at discrete time points. The first of the two approaches matches moments of the Fokker-Planck equation to sample moments of the data, while the second approach employs an Expectation-Maximisation based algorithm using a Monte Carlo bridge sampling scheme to optimise the data likelihood. We derive and numerically test the ability of the two approaches to infer the spatial correlation length of the underlying noise.
Recommendations
- Stochastic metamorphosis with template uncertainties
- String methods for stochastic image and shape matching
- Diffusion bridges for stochastic Hamiltonian systems and shape evolutions
- Extension to infinite dimensions of a stochastic second-order model associated with shape splines
- Shape splines and stochastic shape evolutions: a second order point of view
Cites work
- An integrable shallow water equation with peaked solitons
- Bayesian estimation of discretely observed multi-dimensional diffusion processes using guided proposals
- Bounds for the fundamental solution of a parabolic equation
- Computing large deformation metric mappings via geodesic flows of diffeomorphisms
- Diffeomorphic image matching with left-invariant metrics
- Differential evolution -- a simple and efficient heuristic for global optimization over continuous spaces
- Differential geometry and stochastic dynamics with deep learning numerics
- Existence and uniqueness for stochastic 2D Euler flows with bounded vorticity
- Extension to infinite dimensions of a stochastic second-order model associated with shape splines
- Fundamentals of stochastic filtering
- Geometric dissipation in kinetic equations
- Geometric mechanics. Part I: Dynamics and symmetry.
- Geometry of image registration: the diffeomorphism group and momentum maps
- Guided proposals for simulating multi-dimensional diffusion bridges
- Higher-order spatial accuracy in diffeomorphic image registration
- scientific article; zbMATH DE number 51724 (Why is no real title available?)
- scientific article; zbMATH DE number 3567782 (Why is no real title available?)
- scientific article; zbMATH DE number 1222290 (Why is no real title available?)
- scientific article; zbMATH DE number 1300852 (Why is no real title available?)
- scientific article; zbMATH DE number 523032 (Why is no real title available?)
- scientific article; zbMATH DE number 850878 (Why is no real title available?)
- scientific article; zbMATH DE number 3053873 (Why is no real title available?)
- Importance sampling techniques for estimation of diffusion models
- Langevin equations for landmark image registration with uncertainty
- Moment invariants for the Vlasov equation
- Momentum maps and stochastic Clebsch action principles
- Most likely separation of intensity and warping effects in image registration
- Multiscale turbulence models based on convected fluid microstructure
- Noise and dissipation on coadjoint orbits
- On Transforming a Certain Class of Stochastic Processes by Absolutely Continuous Substitution of Measures
- Orbits of Families of Vector Fields and Integrability of Distributions
- Parametric inference for mixed models defined by stochastic differential equations
- Semiconductor quantum optics.
- Shape Manifolds, Procrustean Metrics, and Complex Projective Spaces
- Shape splines and stochastic shape evolutions: a second order point of view
- Shapes and diffeomorphisms
- Simulation of conditioned diffusion and application to parameter estimation
- Simulation of Multivariate Diffusion Bridges
- Solution properties of a 3D stochastic Euler fluid equation
- Statistical shape analysis. With applications in R
- Stochastic differential equations. An introduction with applications.
- Stochastic discrete Hamiltonian variational integrators
- Stochastic geometric models with non-stationary spatial correlations in Lagrangian fluid flows
- The Euler-Poincaré equations and semidirect products with applications to continuum theories
- The momentum map representation of images
- Towards a Coherent Statistical Framework for Dense Deformable Template Estimation
- Variational principles for stochastic fluid dynamics
- Variational principles for stochastic soliton dynamics
- Variational problems on flows of diffeomorphisms for image matching
Cited in
(32)- Stochastic development regression using method of moments
- Statistics for shape and geometric features. Abstracts from the workshop held July 3--9, 2016
- Stochastic metamorphosis in imaging science
- Statistics for data with geometric structure. Abstracts from the workshop held January 21--27, 2018
- String methods for stochastic image and shape matching
- Random planar shapes and their statistical recognition
- Implications of Kunita-Itô-Wentzell formula for \(k\)-forms in stochastic fluid dynamics
- Continuous-discrete smoothing of diffusions
- Stochastic modeling of geometrical uncertainties on complex domains, with application to additive manufacturing and brain interface geometries
- Networks of coadjoint orbits: from geometric to statistical mechanics
- An infinitesimal probabilistic model for principal component analysis of manifold valued data
- Intrinsic random fields and image deformations
- Extension to infinite dimensions of a stochastic second-order model associated with shape splines
- Statistics of stochastic differential equations on manifolds and stratified spaces. Abstracts from the workshop held October 3--9, 2021 (hybrid meeting)
- Shape splines and stochastic shape evolutions: a second order point of view
- Bayesian statistical shape analysis on the manifold of diffeomorphisms
- Stochastic metamorphosis with template uncertainties
- Simulation of elliptic and hypo-elliptic conditional diffusions
- Langevin equations for landmark image registration with uncertainty
- A Two-Step Geometric Framework For Density Modeling
- Gaussian process landmarking on manifolds
- A CONSISTENT FRAMEWORK FOR A STATISTICAL ANALYSIS OF SURFACES BASED ON GENERALIZED STOCHASTIC PROCESSES
- Diffusion bridges for stochastic Hamiltonian systems and shape evolutions
- New likelihoods for shape analysis
- Random Spatial Structure of Geometric Deformations and Bayesian Nonparametrics
- scientific article; zbMATH DE number 2221430 (Why is no real title available?)
- scientific article; zbMATH DE number 4182527 (Why is no real title available?)
- Conditioning continuous-time Markov processes by guiding
- Most probable flows for Kunita SDEs
- Moment evolution equations and moment matching for stochastic image EPDiff
- Stochastic shape analysis
- Controllability on landmark manifolds for shapes and neural ODEs
This page was built for publication: A geometric framework for stochastic shape analysis
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2420636)
