3D parts decomposition from sparse range data using information criterion

A method which produces a 3-D structural description from contour data or sparse range data by using superquadrics is proposed. The method consists of primal segmentation using superquadrics and convex parts merger using Akaike´s information criterion (AIC) as the criterion. The primal segmentation operation produces many convex parts by expanding superquadrics within the object. The AIC criterion enables keeping the number of parts reasonable because it determines how many parts form the object. An AIC for merging superquadrics is derived. The proposed method is tested successfully by transforming contour data and 3-D sparse range data of human beings into parts descriptions. The tests indicate that the proposed method can be used for 3-D object recognition as well as for data capture for computer graphics applications