An optimal estimation approach to visual perception and learning

How does the visual system learn an internal model of the external environm ent? How is this internal model used during visual perception? How are occl usions and background clutter so effortlessly discounted for when recognizi ng a familiar object? How is a particular object of interest attended to an d recognized in the presence of other objects in the field of view? In this paper, we attempt to address these questions from the perspective of Bayes ian optimal estimation theory. Using the concept of generative models and t he statistical theory of Kalman filtering, we show how static and dynamic e vents occurring in the visual environment may be learned and recognized giv en only the input images. We also describe an extension of the Kalman filte r model that can handle multiple objects in the field of view. The resultin g robust Kalman filter model demonstrates how certain forms of attention ca n be viewed as an emergent property of the interaction between top-down exp ectations and bottom-up signals. Experimental results are provided to help demonstrate the ability of such a model to perform robust segmentation and recognition of objects and image sequences in the presence of occlusions an d clutter. (C) 1999 Elsevier Science Ltd. All rights reserved.