Animate Vision

Dana H. Ballard, Randal C. Nelson, and Brian Yamauchi
Department of Computer Science
University of Rochester
Rochester, NY 14627

Abstract: Animate vision system incorporate directed sensor motion by actively positioning the camera coordinate system in response to physical stimuli. The computations involved in passive vision are so notoriously expensive that one might suspect that the added capability of animation increases the cost. Unexpectedly, experiments with mobile camera platforms are showing just the opposite: visual computation is vastly less expensive when considered in the larger context of behavior.