Abstract
A stereo matching strategy that involves the usage of structural description from the image is proposed. This structural matching strategy is to address the problem of image features that undergo occlusion and also the missing feature situation. The description of the image scene is done by the construction of a relational graph that described the relationship among image primitives. Consequently, the matching problem is to match two structural descriptions, which is represented by a relational graph. The matching between these relational graphs is determined by comparing these structures using graph theory. The best available match between these relational graphs can be determined by finding the best maximal clique in an association graph.