A System for Stereo Computer Vision with Geometric Models

The goal of t h i s work is the s p e c i f i c a t i o n of a system which ex t rac t s 3-D fea tures from images by stereo comparison, and which recognizes ob jec ts by comparing these features to geometric models of ob j ec t s . Models are encoded from hand measurements and cons is t of p iecewisel inear wire frame s t r u c t u r e s . The work is in progress and some p a r t i a l r e s u l t s are shown.