A CONCURRENT, HIERARCHICAL APPROACH TO SYMBOLIC DYNAMIC SCENE INTERPRETATION

Traditional methods for image scene interpretation and understanding a re based mainly on such single threaded procedural paradigms as hypoth esize-and-test or syntactic parsing. As a result, these systems are un able to carry out tasks that require concurrent hypotheses. In this pa per we describe a hierarchical, network-of-frames system for symbolica lly interpreting images. The system is able to interpret a dynamic eve nt based on object motion and interaction among objects. Such capabili ty can be applied to many image applications such as biomedical images , traffic control and behaviour studies. The system has been implement ed in an object-oriented environment in the logic programming language Parlog++ and propagates uncertainty through each frame using Baldwin' s (1986) formulation. The system is illustrated with the legal interpr etation of traffic intersection images. Copyright (C) 1996 Pattern Rec ognition Society.