Two Problems in Analyzing Scenes
Work reported herein was conducted at the Artificial Intelligence Laboratory, a Massachusetts Institute of Technology research program supported in part by the Advanced Research Projects Agency of the Department of Defense and monitored by the Office of Naval Research under Contract Number N00014-70-A-0362-0002.
This paper is based on a B.S. thesis supervised by Patrick Winston. It deals with some previously unexplored problems in the analysis of visual scenes. The scenes consist of two dimensional line drawings of simple objects such as blocks and wedges. The problems have come out of the work that Patrick Winston has done and in discussing them I will be assuming the environment of his system. The first problem asks the questions "When is an object standing? When is it lying?" In the course of answering this question a method is developed for determining the relative true dimensions of an object from its two dimensional oblique projection. The second problem develops methods for discovering when on object is in front of another in situations where previous methods have failed.