Range Image Segmentation

Abstract

The output of a range image sensor is the scene surface geometry in sampled form. It is most commonly represented as an image that is, a two-dimensional array of pixels, each pixel conveying the range of a sampled point on the surface from a reference point. This representation, though voluminous, conveys very little geometric or semantic information about the visible surfaces in the scene. The role of segmentation is to extract geometric primitives relevant to higher level cognitive processes from the pixel-level representation of a range image. The segmentation process partitions a range image into geometric primitives so that all the image pixels are grouped into clusters with a common geometric representation or property.