Why Aspect Graphs Are Not (Yet) Practical for Computer Vision

Abstract

The aspect graph of an object is a graph structure in which each node represents a general view of the object as seen from some maximal, connected cell of viewpoint space. each arc represents an accidental view (or visual event) which occurs on the boundary between two cells of general viewpoint. there is a node for each possible general view of the object. and there is an arc for each possible visual event. The aspect graph representation is often considered to have great potential for computer vision. In the last few years. algorithms have been developed to automatically compute the aspect graphs of polyhedra. general curved objects. and even objects with articulated connections between parts. However, much of the work in this area has a somewhat theoretical flavor and it is not clear that the aspect graph representation, at least as it is currently conceived. will find practical application. For this panel discussion, several distinguished researchers have agreed to provide a critique of the aspect graph approach, and several other distinguished researchers who are working in the aspect graph area have agreed to respond to the critiques. It is hoped that this exchange will help to identify the essential issues involved in this area of research.