Active and Adaptive Vision: Neural Network Models

Abstract

To capture and process visual information flexibly and efficiently from changing external world, the function of active and adaptive information processing is indispensable. Visual information processing in the brain can be interpreted as a process of eliminating irrelevant information from a flood of signals received by the retina. Selective attention is one of the essential mechanisms for this kind of active processing. Selforganization of the neural network is another important function for flexible information processing. This paper introduces some neural network models for these mechanisms from the works of the author: such as “recognition of partially occluded patterns”, “recognition and segmentation of face with selective attention”, “binding form and motion with selective attention” and “self-organization of shift-invariant receptive fields”.