On Locating Objects by Their Distinguishing Features in Multisensory Images
by Tenenbaum, Jay M.
Technical Note 84
Institution: AI Center, SRI International
Address: 333 Ravenswood Ave, Menlo Park, CA 94025
Note: Presented at the Japanese/American Seminar on Picture Processing and Scene Analysis, Kyoto, Japan. SRI Projects 1187 and 1530. This research was jointly supported by the Advanced Research Projects Agency of the Department of Defense (monitored by U.S. Army Research Office– Durham under Grant/Contract No. DAHC04-72-C-0008) and by the Office of Naval Research under Contract NO. N00014-71-C-0294.
This paper reports preliminary work on a knowledge-based perceptual system for a robot that must function in an actual office environment. This system is distinguished by the following pragmatic considerations: 1. It is designed to find specific objects needed by the robot in the performance of a task rather than attempting the usually unnecessary and very much harder job of completely describing an environment of potentially overwhelming complexity. 2. It is based on the premise that in real scenes there is a sufficient redundancy of perceptual clues, as well contextual constraints among objects, so that an intelligent system can devise a relatively simple strategy for distinguishing the specific objects of interest from others likely to be present. 3. It relies heavily on multisensory (i.e., color and range) data to increase the likelihood of finding distinguishing surface attributes for a particular object. Similarly, detailed descriptive representations for complex attributes (e.g., shape and color) are avoided in favor of the simplest representations sufficient to distinguish the object of interest.
|Tenenbaum, J. M.||Alumnus|