INCREMENTAL LEARNING IN HIERARCHICAL NEURAL NETWORKS FOR OBJECT RECOGNITION

Rebecca Fay, Friedhelm Schwenker, Günther Palm

Abstract

Robots that perform non-trivial tasks in real-world environments are likely to encounter objects they have not seen before. Thus the ability to learn new objects is an essential skill for advanced mobile service robots. The model presented in this paper has the ability to learn new objects it is shown during run time. This improves the adaptability of the approach and thus enables the robot to adjust to new situations. The intention is to verify whether and how well hierarchical neural networks are suited for this purpose. The experiments conducted to answer this question showed that the proposed incremental learning approach is applicable for hierarchical neural networks and provides satisfactory classification results.

References

  1. Carpenter, G. A. and Grossberg, S. (2002). Adaptive Resonance Theory. In Arbib, M., editor, The Handbook of Brain Theory and Neural Networks, pages 87-90. MIT Press, Cambridge, 2nd edition.
  2. Coppola, D. M., Purves, H. R., McCoy, A. N., and Purves, D. (1998). The distribution of oriented contours in the real world. Proceedings of the National Academy of Sciences USA, 95(7):4002-4006.
  3. Duda, R. O., Hart, P. E., and Stork, D. G. (2001). Pattern Classification. John Wiley & Sons, New York, 2nd edition.
  4. Fay, R., Kaufmann, U., Schwenker, F., and Palm, G. (2004). Learning object recognition in an neurobotic system. 3rd IWK Workshop SOAVE2004 - SelfOrganization of AdaptiVE behavior, Illmenau, Germany, pages 198- 209.
  5. Freeman, W. T. and Roth, M. (1995). Orientation histograms for hand gesture recognition. In IEEE International Workshop on Automatic Face- and GestureRecognition, pages 296-301, Z ürich, Switzerland.
  6. Grossberg, S. (2000). Adaptive resonance theory. Technical Report TR-2000-024, Center for Adaptive Systems and Department of Cognitive and Neural Science, Boston University.
  7. Knoblauch, A., Fay, R., Kaufmann, U., Markert, H., and Palm, G. (2004). Associating Words to Visually Recognized Objects. In Coradeschi, S. and Saffiotti, A., editors, Anchoring symbols to sensor data. Papers from the AAAI Workshop. Technical Report WS-04-03, pages 10-16. AAAI Press, Menlo Park, California.
  8. Nene, S. A., Nayar, S. K., and Murase, H. (1996). Columbia object image library (coil-20). Technical Report Technical Report CUCS-005-96, Columbia University.
  9. Penrose, R. (1955). A generalized inverse for matrices. Mathematical Proceedings of the Cambridge Philosophical Society, 51:406-413.
  10. Schwenker, F. (2001). Solving multi-class pattern recognition problems with tree structured support vector machines. In Radig, B. and Florczyk, S., editors, Mustererkennung 2001, pages 283-290. Springer.
  11. Schwenker, F., Kestler, H. A., and Palm, G. (2001). Three learning phases for radial-basis-function networks. Neural Networks, 14:439-458.
  12. Simon, S., Schwenker, F., Kestler, H. A., Kraetzschmar, G. K., and Palm, G. (2002). Hierarchical object classification for autonomous mobile robots. In International Conference on Artificial Neural Networks (ICANN), pages 831-836.
  13. Tan, A.-H. (1995). Adaptive resonance associative map. Neural Networks, 8(3):437-446.
Download


Paper Citation


in Harvard Style

Fay R., Schwenker F. and Palm G. (2005). INCREMENTAL LEARNING IN HIERARCHICAL NEURAL NETWORKS FOR OBJECT RECOGNITION . In Proceedings of the Second International Conference on Informatics in Control, Automation and Robotics - Volume 4: ICINCO, ISBN 972-8865-30-9, pages 298-303. DOI: 10.5220/0001182002980303


in Bibtex Style

@conference{icinco05,
author={Rebecca Fay and Friedhelm Schwenker and Günther Palm},
title={INCREMENTAL LEARNING IN HIERARCHICAL NEURAL NETWORKS FOR OBJECT RECOGNITION},
booktitle={Proceedings of the Second International Conference on Informatics in Control, Automation and Robotics - Volume 4: ICINCO,},
year={2005},
pages={298-303},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001182002980303},
isbn={972-8865-30-9},
}


in EndNote Style

TY - CONF
JO - Proceedings of the Second International Conference on Informatics in Control, Automation and Robotics - Volume 4: ICINCO,
TI - INCREMENTAL LEARNING IN HIERARCHICAL NEURAL NETWORKS FOR OBJECT RECOGNITION
SN - 972-8865-30-9
AU - Fay R.
AU - Schwenker F.
AU - Palm G.
PY - 2005
SP - 298
EP - 303
DO - 10.5220/0001182002980303