Accumulated Visual Representation for Cognitive Vision

Pugeault, N. , Wörgötter, F. and Krüger, N. (2008) Accumulated Visual Representation for Cognitive Vision. In: British Machine Vision Conference 2008, Leeds, UK, 01-04 Sep 2008, ISBN 9781901725360 (doi: 10.5244/C.22.98)

Full text not currently available from Enlighten.

Abstract

In this paper we present a scheme for accumulating local visual informa- tion in 3D, under known motion. Information about the object’s 3D shape is provided by reconstructing local contour descriptors. This shape informa- tion is accumulated over time in three ways: 1) disambiguation: erroneous stereo correspondences that are unsuccessfully tracked are discarded. We make use of aspect cues to increase the data association selectivity. 2) cor- rection: the full pose of the reconstructed features is corrected over time using an Kalman Filter approach. 3) completeness: multiple 2 1 D represen- 2 tations become merged, constructing a full 3D representation of the object. The described system is evaluated quantitatively on three different scenarios.

Item Type:Conference Proceedings
Status:Published
Refereed:Yes
Glasgow Author(s) Enlighten ID:Pugeault, Dr Nicolas
Authors: Pugeault, N., Wörgötter, F., and Krüger, N.
College/School:College of Science and Engineering > School of Computing Science
ISBN:9781901725360

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