Proceedings

Incremental Learning of Perceptual Categories for Open-Domain Sketch Recognition

Download BibTeX Cite

Proceedings of the Twentieth International Joint Conference on Artificial Intelligence. Pages 447-451.

© 2007 International Joint Conferences on Artificial Intelligence

Abstract

Most existing sketch understanding systems require a closed domain to achieve recognition. This paper describes an incremental learning technique for open-domain recognition. Our system builds generalizations for categories of objects based upon previous sketches of those objects and uses those generalizations to classify new sketches. We represent sketches qualitatively because we believe qualitative information provides a level of description that abstracts away details that distract from classification, such as exact dimensions. Bayesian reasoning is used in building representations to deal with the inherent uncertainty in perception. Qualitative representations are compared using SME, a computational model of analogy and similarity that is supported by psychological evidence, including studies of perceptual similarity. We use SEQL to produce generalizations based on the common structure found by SME in different sketches of the same object. We report on the results of testing the system on a corpus of sketches of everyday objects, drawn by ten different people.

  1. [Alvarado et al., 2002] Alvarado, C., Oltmans, M., and Davis, R. A framework for multi-domain sketch recogni- tion. In 2002 AAAI Spring Symposium on Sketch Under- standing, Palo Alto, CA, 2002.
  2. [Alvarado and Davis, 2005] Alvarado, C., and Davis, R. Dynamically constructed Bayes nets for multi-domain sketch understanding. In Proceedings of the 19 th Interna- tional Joint Conference on Artificial Intelligence, pages 1407-1412, Edinbourgh, Scotland, 2005. IJCAI
  3. [Anderson et al., 2004] Anderson, D., Bailey, C., and Sku- bic, M. Hidden Markov model symbol recognition for sketch-based interfaces. In Making Pen-Based Interaction Intelligent and Natural, pages 15-21, Arlington, VA, 2004. AAAI Press.
  4. [Bokor and Ferguson, 2004] Bokor, J. L., and Ferguson, R. W. Integrating probabilistic reasoning into a symbolic diagrammatic reasoner. In Proceedings of the 18 th Inter- national Workshop on Qualitative Reasoning (QR'04), Evanston, IL, 2004.
  5. [Buckley, 1979] Buckley, S. Sun Up to Sun Down. McGraw Hill, New York, 1979 [
  6. [Falkenhainer et al., 1989] Falkenhainer, B., Forbus, K. and Gentner, D. The Structure-Mapping Engine: Algorithms and examples. Artificial Intelligence, 41: 1-63, 1989.
  7. [Ferguson and Forbus, 1999] Ferguson, R. W., and Forbus, K. D. GeoRep: A flexible tool for spatial representations of line drawings. In Proceedings of the 13 th International Workshop on Qualitative Reasoning (QR'99), pages 84- 91. Loch Awe, Scotland, 1999.
  8. [Forbus et al., 2004] Forbus, K., Lockwood, K., Klenk, M., Tomai, E., and Usher, J. Open-domain sketch understand- ing: The nuSketch approach. In AAAI Fall Symposium on Making Pen-based Interaction Intelligent and Natural, pages 58-63, Washington, DC, 2004. AAAI Press.
  9. [Gentner, 1983] Gentner, D. Structure-mapping: A theoreti- cal framework for analogy. Cognitive Science, 7: 155- 170, 1983.
  10. [Halstead and Forbus, 2005] Halstead, D., and Forbus, K. Transforming between propositions and features: Bridg- ing the gap. In Proceedings of the 20 th National Confer- ence on Artificial Intelligence (AAAI'05), pages 777-782, Pittsburgh, PA, 2005. AAAI Press.
  11. [Hoffman and Richards, 1984] Hoffman, D. D., and Rich- ards, W. A. Parts of recognition. Cognition, 18: 65-96, 1984.
  12. [Kuehne et al., 2000] Kuehne, S., Forbus, K., Gentner, D. and Quinn, B. SEQL: Category learning as progressive abstraction using structure mapping. In Proceedings of the 22 nd Annual Conference of the Cognitive Science Society, pages 770-775, Philadelphia, PA, 2000.
  13. [Liwicki and Knipping, 2005] Liwicki, M., and Knipping, L. Recognizing and simulating sketched logic circuits. In Proceedings of the 9 th International Conference on Knowledge-Based Intelligent Information & Engineering Systems, pages 588 -594, Melbourne, Australia, 2005.
  14. [Lowe, 1989] Lowe, D. G. Organization of smooth image curves at multiple scales. International Journal of Com- puter Vision, 3(2): 119-130, 1989.
  15. [Marr, 1982] Marr, D. Vision. W.H. Freeman and Company, New York, 1982.
  16. [Museros and Escrig, 2004] Museros, L., & Escrig, M. T. 2004. A qualitative theory for shape representations and matching. In Proceedings of the 18 th International Work- shop on Qualitative Reasoning (QR'04), Evanston, IL, 2004.
  17. [Park and Kwon, 2003] Park, J., and Kwon, Y-B. Main wall recognition of architectural drawings using dimension ex- tension line. In Proceedings of the Fifth IAPR Interna- tional Workshop on Graphics Recognition (GREC'03), pages 116-127, Barcelona, Spain, 2003. Springer.
  18. [Veselova and Davis, 2004] Veselova, O., and Davis, R. Perceptually based learning of shape descriptions. In Pro- ceedings of the 19 th National Conference on Artificial In- telligence (AAAI'04), pages 482-487, San Jose, CA, 2004. AAAI Press.

References were extracted automatically from the PDF and may contain errors.

Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards