Share Email Print
cover

Proceedings Paper

Object recognition by active fusion
Author(s): Manfred Prantl; Hermann Kopp-Borotschnig; Harald Ganster; David Sinclair; Axel J. Pinz
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Today's computer vision applications often have to deal with multiple, uncertain, and incomplete visual information. In this paper, we apply a new method, termed 'active fusion', to the problem of generic object recognition. Active fusion provides a common framework for active selection and combination of information from multiple sources in order to arrive at a reliable result at reasonable costs. In our experimental setup we use a camera mounted on a 2m by 1.5m x/z-table observing objects placed on a rotating table. Zoom, pan, tilt, and aperture setting of the camera can be controlled by the system. We follow a part-based approach, trying to decompose objects into parts, which are modeled as geons. The active fusion system starts from an initial view of the objects placed on the table and is continuously trying to refine its current object hypotheses by requesting additional views. The implementation of active fusion on the basis of probability theory, Dempster-Shafer's theory of evidence and fuzzy set theory is discussed. First results demonstrating segmentation improvements by active fusion are presented.

Paper Details

Date Published: 29 October 1996
PDF: 11 pages
Proc. SPIE 2904, Intelligent Robots and Computer Vision XV: Algorithms, Techniques,Active Vision, and Materials Handling, (29 October 1996); doi: 10.1117/12.256290
Show Author Affiliations
Manfred Prantl, Technical Univ. of Graz (Austria)
Hermann Kopp-Borotschnig, Technical Univ. of Graz (Austria)
Harald Ganster, Technical Univ. of Graz (Austria)
David Sinclair, Technical Univ. of Graz (Austria)
Axel J. Pinz, Technical Univ. of Graz (Austria)


Published in SPIE Proceedings Vol. 2904:
Intelligent Robots and Computer Vision XV: Algorithms, Techniques,Active Vision, and Materials Handling
David P. Casasent, Editor(s)

© SPIE. Terms of Use
Back to Top