A LATTICE FRAMEWORK FOR INTEGRATING VISION MODULES

Because of the successes in understanding information processing by individual modules such as stereopsis, motion, texture and color, research in computational vision is now turning to studies of how information provided by these modules may be integrated or ''assimilated.'' A framework for assimilation based upon a partial ordering of constraints implicit in all active modules is proposed. Such constraints, for example the rigidity constraint for motion, although often robust are also fallible, and hence are more properly regarded as premises. Such premises are used to construct a preference ordering for (classes of) interpretations of the image. Interpretations associated with maximal states in the ordering are taken as the assimilated interpretation of the modules. This approach stresses the need to use world knowledge to reason about the plausibility and consistency of interpretations of the image data.