Abstract

This tutorial provides an understanding of current and past attempts to model vision using simple computing elements arranged as a network. The models examined include the highly publicized parallel distributed processing models described by Rumelhart and McClelland, the work of Stephen Grossberg, and others. The primary goal is to taxonomize, explain, and evaluate these models, measuring their capabilities against long-standing problems in vision, visual development, and computer vision.

© 1988 Optical Society of America