It is suggested that recent advances in the construction of artificial vision systems provide the beginnings of a framework for an information processing analysis of human visual perception. A review is made of some pertinent investigations which have appeared in the psychological literature, along with some of the salient and potentially useful theoretical concepts which have resulted from the attempts to build computer vision systems. An attempt is made to integrate these two sources of ideas to suggest some desirable structural and behavioral concepts which apply to both the natural and the artificial systems.