Journal of Mathematical Imaging and Vision Компьютерное зрение и распознавание образов

Enhancing a Kernel Method for Shape Analysis in Kendall Space

Ximo Gual-Arnau, Juan Monterde
Springer Nature · 2025-03-20 · DOI 10.1007/s10851-025-01235-z

Аннотация

Abstract In learning methods defined on a manifold, it is often important to embed the manifold into a Hilbert space to apply kernel methods, as it enables the utilization of linear techniques in the transformed space, thereby facilitating complex data analysis and classification tasks with nonlinear structures [10]. In [6], the authors define an embedding of the $$\textit{m}$$ m -dimensional Kendall shape space into an $$\mathbb {R}^{\textit{N}}$$ R N space, from which kernel methods are introduced in the Kendall shape space. In this paper, we significantly simplify this result by achieving an embedding of the Kendall shape space into a Euclidean space $$\mathbb {R}^{\textit{n}}$$ R n of much lower dimension $$\textit{n}$$ n than $$\textit{N}$$ N , greatly simplifying calculations and computational costs of applications. Additionally, we characterize the image of the Kendall shape space in $$\mathbb {R}^{\textit{n}}$$ R n , allowing us to define a new extrinsic mean of shapes in the Kendall shape space.

Читать у издателя →