A Hybrid Approach to Range Image Segmentation Based on Differential Geometry

概要

論文の詳細を見る
One of the most significant problems arising out of understanding range images of 3D objects is image segmentation. This paper describes a hybrid approach to problem, where "hybrid" refers to a combination of both region-and edge-based considerations. It is assumed that the range image of objects, which may be constructed of both curved and planar surfaces, is divided into regions. These are meant to correspond to surface primitives that are homogeneous in their intrinsic differential geometric properties and do not contain discontinuities in either depth or surface orientation. The method is based on the computation of first and second partial derivatives, which are obtained by locally approximating object surfaces, using biquadratic polynomials. By computing the Gaussian and mean curvatures and examining their signs, an initial region-based segmentation is then obtained in the form of a curvature sign map. Two initial edge-based segmentations are also computed from the partial derivatives and depth values. One detects jump edges while the other highlights roof edges. The three image maps are then combined to produce the final segmentation. Experimental results are presented for both synthetic and real range data. These indicate that the proposed segmentation method is useful for describing both polyhedral and curved objects.
一般社団法人情報処理学会の論文
1992-03-15