1,721,069 research outputs found
Dynamic shape analysis and comparison of leaf growth
No full textIn the statistical analysis of shape a goal beyond the analysis of static shapes lies in the quantification of `same' deformation of different shapes. Typically, shape spaces are modelled as Riemannian manifolds on which parallel transport along geodesics naturally qualifies as a measure for the `similarity' of deformation. Since these spaces are usually defined as combinations of Riemannian immersions and submersions, only for few well featured spaces such as spheres or complex projective spaces (which are Kendall's spaces for 2D shapes), parallel transport along geodesics can be computed explicitly. In this contribution a general numerical method to compute parallel transport along geodesics when no explicit formula is available is provided. This method is applied to the shape spaces of closed 2D contours based on angular direction and to Kendall's spaces of shapes of arbitrary dimension. In application to the temporal evolution of leaf shape over a growing period, one leaf's shape-growth dynamics can be applied to another leaf. For a specific poplar tree investigated it is found that leaves of initially and terminally different shape evolve rather parallel, i.e. with comparable dynamics
A comment to “Statistics on manifolds and landmark based image analysis: A nonparametric theory with applications”
No full textEin Extremalproblem für das harmonische Maß einer Familie von Extremalkontinua im Einheitskreis
No full textEin Extremalproblem für das harmonische Maß einer Familie von Extremalkontinua im Einheitskreis
No full textManifold stability and the central limit theorem for mean shape
No full textIn the literarature there is an overwhelming number of concepts for a “mean” on shape spaces. First we classify those into three fundamental types. Then we introduce manifold stability which is essential to a apply a standard two sample test on a non-manifold shape space. For sample means on Kendall’s shape spaces, we give the proof to establish manifold stability and illustrate consequences for the discrimination of non-concentrated shapes. We conclude with a discussion of the result for general shape spaces and population means
Manifold stability and the central limit theorem for mean shape
No full textIn the literarature there is an overwhelming number of concepts for a “mean” on shape spaces. First we classify those into three fundamental types. Then we introduce manifold stability which is essential to a apply a standard two sample test on a non-manifold shape space. For sample means on Kendall’s shape spaces, we give the proof to establish manifold stability and illustrate consequences for the discrimination of non-concentrated shapes. We conclude with a discussion of the result for general shape spaces and population means
- …
