Dual-coding representations for robot vision in Tekkotsu - Robotics Institute Carnegie Mellon University

Dual-coding representations for robot vision in Tekkotsu

David S. Touretzky, N. S. Halelamien, Ethan Tira-Thompson, J. J. Wales, and Kei Usui
Journal Article, Autonomous Robots, Vol. 22, No. 4, pp. 425 - 435, May, 2007

Abstract

We describe complementary iconic and symbolic representations for parsing the visual world. The iconic pixmap representation is operated on by an extensible set of ?isual routines?(Ullman, 1984; Forbus et al., 2001). A symbolic representation, in terms of lines, ellipses, blobs, etc., is extracted from the iconic encoding, manipulated algebraically, and re-rendered iconically. The two representations are therefore duals, and iconic operations can be freely intermixed with symbolic ones. The dual-coding approach offers robot programmers a versatile collection of primitives from which to construct application-specific vision software. We describe some sample applications implemented on the Sony AIBO.

BibTeX

@article{Touretzky-2007-9648,
author = {David S. Touretzky and N. S. Halelamien and Ethan Tira-Thompson and J. J. Wales and Kei Usui},
title = {Dual-coding representations for robot vision in Tekkotsu},
journal = {Autonomous Robots},
year = {2007},
month = {May},
volume = {22},
number = {4},
pages = {425 - 435},
}