Towards binocular active vision in a robot head system

Haitham, F., Aragon-Camarasa, G. and Siebert, J.P. (2008) Towards binocular active vision in a robot head system. In: Towards Autonomous Robotic Systems Conference, Edinburgh, 1-3 September,

[img] Text
502x.pdf

953kB

Abstract

This paper presents the first results of an investigation and pilot study into an active, binocular vision system that combines binocular vergence, object recognition and attention control in a unified framework. The prototype developed is capable of identifying, targeting, verging on and recognizing objects in a highly-cluttered scene without the need for calibration or other knowledge of the camera geometry. This is achieved by implementing all image analysis in a symbolic space without creating explicit pixel-space maps. The system structure is based on the ‘searchlight metaphor’ of biological systems. We present results of a first pilot investigation that yield a maximum vergence error of 6.4 pixels, while seven of nine known objects were recognized in a high-cluttered environment. Finally a “stepping stone” visual search strategy was demonstrated, taking a total of 40 saccades to find two known objects in the workspace, neither of which appeared simultaneously within the Field of View resulting from any individual saccade.

Item Type:Conference Proceedings
Keywords:Binocular vision, active vision, gaze control, vergence, robot head
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Siebert, Dr Jan and Aragon Camarasa, Dr Gerardo
Authors: Haitham, F., Aragon-Camarasa, G., and Siebert, J.P.
Subjects:T Technology > T Technology (General)
Q Science > QA Mathematics > QA75 Electronic computers. Computer science
College/School:College of Science and Engineering > School of Computing Science
First Published:First published in Proceedings of the Towards Autonomous Robotic Systems Conference
Publisher Policy:Reproduced with permission of the author

University Staff: Request a correction | Enlighten Editors: Update this record