Extraction of Reliable Primitives from Unorganized Point Clouds

Abstract

Manufactured objects are the combination of some basic geometric primitives such as planes, spheres, cylinders, etc. Existing methods often approximate data points by geometric primitives for surface reconstruction. In this context, the extraction of reliable geometric primitives from unorganized point clouds is addressed in this paper. First, curved surfaces (cylinders and spheres) are extracted robustly from point clouds. Then the points associated with these primitives are removed. Finally, planar surfaces are extracted from the remaining points using a RANSAC-based approach. The performance of the proposed framework is tested on synthetic and scanned data. The performance analysis shows that the proposed method outperforms existing methods and may be used for various applications.