Multi-view object instance recognition in an industrial context

Mustafa, W., Pugeault, N. , Buch, A. G. and Krüger, N. (2017) Multi-view object instance recognition in an industrial context. Robotica, 35(2), pp. 271-292. (doi: 10.1017/s0263574715000430)

Full text not currently available from Enlighten.

Abstract

We present a fast object recognition system coding shape by viewpoint invariant geometric relations and appearance information. In our advanced industrial work-cell, the system can observe the work space of the robot by three pairs of Kinect and stereo cameras allowing for reliable and complete object information. From these sensors, we derive global viewpoint invariant shape features and robust color features making use of color normalization techniques. We show that in such a set-up, our system can achieve high performance already with a very low number of training samples, which is crucial for user acceptance and that the use of multiple views is crucial for performance. This indicates that our approach can be used in controlled but realistic industrial contexts that require—besides high reliability—fast processing and an intuitive and easy use at the end-user side.

Item Type:Articles
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Pugeault, Dr Nicolas
Authors: Mustafa, W., Pugeault, N., Buch, A. G., and Krüger, N.
College/School:College of Science and Engineering > School of Computing Science
Journal Name:Robotica
Publisher:Cambridge University Press
ISSN:0263-5747
ISSN (Online):1469-8668
Published Online:23 June 2015

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