Shape splines and stochastic shape evolutions: a second order point of view

Hamiltonian systems on groups of diffeomorphisms and on manifolds of mappings and metrics (37K65) Optimization of shapes other than minimal surfaces (49Q10) Geodesic flows in symplectic geometry and contact geometry (53D25) Stochastic ordinary differential equations (aspects of stochastic analysis) (60H10) Numerical solutions to stochastic differential and integral equations (65C30) Numerical computation using splines (65D07) Numerical optimization and variational techniques (65K10)

Abstract: This article presents a new mathematical framework to perform statistical analysis on time-indexed sequences of 2D or 3D shapes. At the core of this statistical analysis is the task of time interpolation of such data. Current models in use can be compared to linear interpolation for one dimensional data. We develop a spline interpolation method which is directly related to cubic splines on a Riemannian manifold. Our strategy consists of introducing a control variable on the Hamiltonian equations of the geodesics. Motivated by statistical modeling of spatiotemporal data, we also design a stochastic model to deal with random shape evolutions. This model is closely related to the spline model since the control variable previously introduced is set as a random force perturbing the evolution. Although we focus on the finite dimensional case of landmarks, our models can be extended to infinite dimensional shape spaces, and they provide a first step for a non parametric growth model for shapes taking advantage of the widely developed framework of large deformations by diffeomorphisms.

Review text

This paper proposes a second-order method for the shape evolution analysis. For this purpose the authors introduce the Hamiltonian equations of geodesics for landmark matching. Then an external force is added to the landmark configuration. Assuming a random distribution of this additional term, the authors derive a stochastic differential equation as model and use it to obtain a new minimization problem for the growth estimation to obtain a corresponding spline as description of the shape evolution. After proving existence results for this new model and further theoretical properties, the authors present several numerical studies also considering noisy data as input for the reconstruction of the shape evolution. Throughout, the authors concentrate on the finite-dimensional case but discuss the extension to the infinite-dimensional setting.

Recommendations

Cites work

Cited in

(26)

Describes a project that uses

Uses Software

LDDMM

This page was built for publication: Shape splines and stochastic shape evolutions: a second order point of view

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2892167)