Abstract

We present a novel algorithm for extracting shapes of con- tours of (possibly partially occluded) ob jects from noisy or low-contrast images. The approach taken is Bayesian: we adopt a region-based model that incorporates prior knowledge of specific shapes of interest. To quan- tify this prior knowledge, we address the problem of learning probability models for collections of observed shapes. Our method is based on the geometric representation and algorithmic analysis of planar shapes in- troduced and developed in [15]. In contrast with the commonly used approach to active contours using partial differential equation methods [12,20,1], we model the dynamics of contours on vector fields on shape manifolds.