FINDING PLANES AND CLUSTERS OF OBJECTS FROM 3D LINE SEGMENTS WITH APPLICATION TO 3D MOTION DETERMINATION

We address in this paper how to find clusters based on proximity and p lanar facets based on coplanarity from 3D line segments obtained from stereo. The proposed methods are efficient and have been tested with m any real stereo data. These procedures are indispensable in many appli cations including scene interpretation, object modeling, and object re cognition. We show their application to 3D motion determination. We ha ve developed an algorithm based on the hypothesize-and-verify paradigm to register two consecutive 3D frames obtained from stereo and estima te their transformation/motion. By grouping 3D line segments in each f rame into clusters and planes, we can reduce effectively the complexit y of the hypothesis generation phase. (C) 1994 Academic Press, Inc.