Skip to main content
SpringerLink
Log in

Ein hybrides Bildanalyse-System für einen künstlichen Kommunikator

A Hybrid Vision Frontend for an Artificial Communicator

  • Published:
Kognitionswissenschaft

Summary

In this contribution, we present the “visual system” of an artificial communicator, which enables the communicator to recognize objects (wooden toy pieces) in camera images. It is based on a hybrid approach, which applies artificial neural nets for a holistic representation of low level knowledge. The transition to the symbolic level is realized using a semantic network as a knowledge base containing explicit object models. In the next processing step, the information extracted from single images is integrated by a scene memory to a representation suitable for the artificial communicator. On the one hand, the memory stabilizes the data extracted from static scenes, on the other hand it realizes an efficient representation of changing scenes calculating the “difference”. By these means, other modules of the communicator working on different time scales can access the scene information interactively, which is the prerequisite for a dialogue with the user.

Zusammenfassung

In diesem Beitrag wird das “Sehsystem” eines künstlichen Kommunikators vorgestellt, das die Erkennung von Objekten (Holzspielzeug) in Kamerabildern ermöglicht. Es basiert auf einem hybriden Ansatz, der künstliche neuronale Netze zur holistischen Repräsentation signalnahen Wissens einsetzt. Der Übergang zur symbolischen Ebene erfolgt durch die Ankopplung einer Wissens-basis, die auf einer expliziten Objektmodellierung durch ein semantisches Netzwerk beruht. In der nächsten Verarbeitungsstufe werden die aus Einzelbildern gewonnenen Informationen durch ein Szenengedächtnis zu einer für den künstlichen Kommunikator geeigneten Repräsentation integriert. Das Gedächtnis erreicht zum einen eine Stabilisierung der aus statischen Szenen gewonnenen Daten, zum anderen durch Bildung einer “Differenz” eine effiziente Speicherung sich verändernder Szeneninhalte. Damit wird anderen, auf verschiedenen Zeitskalen arbeitenden Modulen des Kommunikators eine interaktive Abfrage der Szeneninformation ermöglicht, was die Voraussetzung für einen Dialog mit dem Benutzer darstellt.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Literatur

  • Agre, P. & Chapman, D. (1990). What Are Plans For? In Maes, P. (Ed.), Designing Autonomous Agents: Theory and Practice from Biology to Engineering and Back, volume 6,1 of Robotics and autonomous systems, pp. 17–34.

  • Ahmad, S. (1991). VISIT: An efficient computational model of human visual attention. PhD thesis, ICSI, Univ. of California, Berkley.

    Google Scholar 

  • Alexander, I. (1989). Neural Computing Architectures. North Oxford Academic Pub. Ltd.

  • Almássy, N. & Verschure, P. (1992). Optimizing self-organization control architectures with genetic algorithms: The interaction between natural selection and ontogenesis. In R. Manderick, B. M. (Ed.), Parallel Problem Solving from Nature, 2: Proceedings of the Second Conference on Parallel Problem Solving from nature, Brussels, pp. 451–460. North-Holland.

  • Blanchet, P. (1994). An Architecture for Representing and Learning Behaviors by Trial and Error. In Proceedings Congress SAB’94: From Animals to Animats III, Simulation of Adaptative Behavior, Brighton (England).

  • Bollmann, M., Hoischen, R. & Mertsching, B. (1997). Interaction of static and dynamic scene — features guiding visual attention. In Paulus, E. & Wahl, F. (Hrsg.), DAGM Mustererkennung 1997, Informatik aktuell, pp. 483–490, Berlin: Springer-Verlag.

    Google Scholar 

  • Brooks, R. (1986). A robust layered control system for a mobile robot. In IEEE Journal of Robotics and Automation, volume 2, 1, pp. 14–23.

    Article  MathSciNet  Google Scholar 

  • Brooks, R. (1991). Challenges for Complete Creature Architectures. In Meyer, J.-A. & Wilson, S. (Eds.), From animals to animats, First International Conference on Simulation of Adaptive Behavior, pp. 434–443.

  • Cameron, S., Grossberg, S. & Guenther, F. H. (1998). A self-organizing neural network architecture for navigation using optic flow. Neural Computation, 10(2):313–352.

    Article  Google Scholar 

  • Chapman, D. (1991). Vision, Instruction & Action. Cambridge, MA: MIT Press

    Google Scholar 

  • Clark, A. (Ed.) (1997). Being There. Putting Brain, Body, and World Together Again. Cambridge, Massachusetts: MIT Press, A Bradford Book

  • Doya, K. (1997). Efficient nonlinear control with actor-tutor architecture. In Mozer, M. C., Jordan, M. I. & Petsche, T. (Eds.), Advances in Neural Information Processing Systems, volume 9, p. 1012. The MIT Press.

  • Fiehler, R. (Ed.) (1980). Kommunikation und Kooperation. Theoretische und empirische Untersuchungen zur kommunikativen Organisation kooperativer Prozesse. Berlin: Einhorn

  • Fislage, M. (1998). Visuelle Aufmerksamkeitssteuerung bei gestikbasierter Mensch-Maschine Interaktion. Master’s thesis, Universität Bielefeld, Technische Fakultät.

  • Franklin, S. (1995). Artificial Minds. Cambridge, MA: MIT Press

    Google Scholar 

  • Franklin, S. & Graesser, A. (1996). Is It an Agent, or Just a Program: A Taxonomy for Autonomous Agents. In Müller, J. P., Woolridge, M. J. & Jennings, N. R. (Eds.), Intelligent Agents III. Agent Theories, Architectures and Languages, pp. 21–35. Berlin: Springer

    Google Scholar 

  • Goecke, K. U. & Milde, J.-T. (1998). Natural language generation in a behavior-oriented robot control architecture. Submitted to publication.

  • Happel, B. L. M. & Murre, J. M. J. (1994). Design and evolution of modular neural-network architectures. Neural Networks, 7(6–7):985–1004.

    Article  Google Scholar 

  • Jackendoff, R. (1990). Semantic Structures. Current studies in linguistics series, 18. Cambridge, MA: MIT Press

    Google Scholar 

  • Jacobs, R. A., Jordan, M. I. & Barto, A. G. (1991). Task decomposition through competition in a modular connectionist architecture — the what and where vision tasks. Cognitive Science, 15(2):219–250.

    Article  Google Scholar 

  • Lobin, H. (1998). Handlungsanweisungen. Sprachliche Spezifikation teilautonomer Aktivität. Gütersloh: Deutscher Universitäts Verlag.

    Google Scholar 

  • Meier, C. & Rieser, H. (1995). Modelling Situated “Reference Shiftsïn Task-Oriented Dialogue. Technical Report 95/12, SFB 360, Universität Bielefeld.

  • Milde, J.-T. (1998). Action-centered communication with an embedded agent. In FLAIRS, Special Track on natural language processing and Human Computer Interaction.

  • Milde, J.-T., Peters, K. & Strippgen, S. (1997). Situated communication with robots. In Proceedings of the first international Workshop on Human-Computer Conversation, Bellagio, Italy.

  • Millikan, R. (Ed.) (1984). Language, Thought, and Other Biological Categories. Cambridge, Massachusetts: MIT Press

  • Niebur, E. & Koch, C. (1996). Control of selective visual attention: Modelling the “where” pathway. In Touretzky, D., Mozer, M. & Hasselmo, M. (Eds.), Advances in Neural Information Processing Systems 8 NIPS*95. Bradford: MIT Press.

    Google Scholar 

  • Nölker, C. & Ritter, H. (1998). Grefit: Visuelle Erkennung Kontinuierlicher Handposturen. In Dassow, J. & Kruse, R. (Hrsg.), Informatik ’98: Informatik zwischen Bild und Sprache, pp. 213–222, Magdeburg: Springer Verlag. Reihe “Informatik aktuell”.

    Google Scholar 

  • Peters, K., Strippgen, S. & Milde, J.-T. (1997). Language processing in action-centered communication. In Tagungsband der DGFS, Sektion Computerlinguistik, Heidelberg.

  • Peters, K., Strippgen, S. & Milde, J.-T. (1998). Cora — an instructable agent. In Proceedings of DARS, Karlsruhe.

  • Rae, R. & Ritter, H. (1998). 3d real-time tracking of points of interest based on zero-disparity filtering. In Posch, S. & Ritter, H. (Hrsg.), Workshop Dynamische Perzeption, Proceedings in Artificial Intelligence, pp. 105–111, Workshop der GI-Fachgruppe 1.0.4 Bildverstehen in Bielefeld, 18.–19. Jun., Germany. Sankt Augustin, Deutschland: INFIX-Verlag

  • Rieser, H. (1996). Repräsentations-Metonymie, Perspektive und aufgabenorientierten Dialogen. Technical Report 96/6, SFB 360, Universität Bielefeld.

  • Steels, L. (1998). The origin of syntax in visually grounded agents. In Artificial Intelligence, pp. 133–156.

  • Torrance, M. (1994). Natural Communication with Robots. Master’s thesis, Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology.

  • van Dam, J. W. M., Kröse, B. J. A. & Groen, F. C. A. (1994). CNN: A neural architecture that learns multiple transformations of spatial representations. In Marinaro, M. & Morasso, P. G. (Eds.), Artificial Neural Networks, pp. 1420–1423. Berlin Heidelberg New York: Springer-Verlag.

    Google Scholar 

  • Weiskrantz, L. (Ed.) (1988). Thought without language. Oxford University Press.

  • Zadel, S. (1996). An algorithm for bootstrapping the core of a biologically inspired motor control system. In von der Malsburg, C., von Seelen, W., Vorbrüggen, J. C. & Sendhoff, B. (Eds.), Proceedings of the International Conference on Artificial Neural Networks (ICANN’96), Bochum, Germany, July 16–19, volume 1112 of Lecture Notes in Computer Science, pp. 629–634, Berlin, Heidelberg, New York: Springer-Verlag.

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. SFB 360, Universität Bielefeld, Postfach 100131, D-33501, Bielefeld

    Gunther Heidemann, Nils Jungclaus, Franz Kummert, Gerhard Sagerer & Helge Ritter

Authors
  1. Gunther Heidemann
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Nils Jungclaus
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Franz Kummert
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Gerhard Sagerer
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Helge Ritter
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Gunther Heidemann.

Additional information

Diese Arbeit wurde von der Deutschen Forschungsgemeinschaft im Rahmen des Sonderforschungsbereichs 360 gefördert.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Heidemann, G., Jungclaus, N., Kummert, F. et al. Ein hybrides Bildanalyse-System für einen künstlichen Kommunikator. Kognit. Wiss. 8, 101–107 (1999). https://doi.org/10.1007/s001970050078

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s001970050078

Keywords

Search

Navigation