VISION CONSTRAINT RECOGNITION SYSTEM: ANALYSING THE ROLE OF REASONING IN HIGH LEVEL VISION.

A model-based vision system, the Visual Constraint Recognition System (VICTORS), which incorporates a truth maintenance system (TMS) as its reasoning tool and has many advantages over existing model-based vision systems, is described. VICTORS can simultaneously identify all occurrence of a given figure within a scene of randomly overlapping rectangles, subject to variable figure geometry, input data from multiple sources (e. g. stereo data) and incomplete figures. Moreover, it conducts sensitivity analyses of figures, updates figures (given new input data) without having to entirely recompute the new figures, and is robust given noise and occlusion.