Data Driven Generation of Interactions for Feature Binding and Relaxation Labeling (original) (raw)

Abstract

We present a combination of unsupervised and supervised learning to generate a compatibility interaction for feature binding and labeling problems. We focus on the unsupervised data driven generation of prototypic basis interactions by means of clustering of proximity vectors, which are computed from pairs of data in the training set. Subsequently a supervised method recently introduced in [9] is used to determine coefficients to form a linear combination of the basis functions, which then serves as interaction. As special labeling dynamic we use the competitive layer model, a recurrent neural network with linear threshold neurons, and show an application to cell segmentation.

Preview

Unable to display preview. Download preview PDF.

Similar content being viewed by others

References

  1. Hahnloser, R., Sarpeshkar, R., Mahowald, M.A., Douglas, R.J., Seung, H.S.: Digital selection and analogue amplification coexist in a cortex-inspired silicon circuit, Nature 405 (2000) 947–951.
    Article Google Scholar
  2. Heidemann, G., Ritter, H.: Efficient Vector Quantization using the WTA-rule with Activity Equalization. Neural Processing Letters vol. 13, no. 1 (2001) 17–30.
    Article MATH Google Scholar
  3. Hofmann, T., Puzicha, J., Buhmann, J.: Unsupervised texture segmentation in a deterministic annealing framework. IEEE Trans. PAMI 20(8) (1998) 803–818.
    Google Scholar
  4. Nattkemper, T.W., Wersing, H., Schubert, W., Ritter, H.: Fluorescence Micrograph Segmentation by Gestalt-Based Feature Binding. Proc. of the Int. Joint Conf. on Neur. Netw. (IJCNN), vol. 1(2000) 248–254, Como, Italy.
    Google Scholar
  5. Pelillo, M., Refice, M.: Learning compatibility coefficients for relaxation labeling processes. IEEE Trans. PAMI, 16(9) (1994) 933–945.
    Google Scholar
  6. Pelillo, M., Abbattista, F., Maffione, A.: An evolutionary approach of training relaxation labeling processes. Pattern Recognition Letters, 16, (1995), 1069–1078.
    Article Google Scholar
  7. Pelillo, M., Fanelli, A.: Autoassociative learning in relaxation labeling networks. Pattern Recognition Letters, 18, (1997), 3–12.
    Article MATH Google Scholar
  8. Rosenfeld, A., Hummel, R., Zucker, S.W.: Scene labeling by relaxation operations. IEEE Trans. Systems, Man Cybernetics 6(6) (1976) 420–433.
    Article MATH MathSciNet Google Scholar
  9. Wersing, H.: Learning Lateral Interactions for Feature Binding and Sensory Segmentation. NIPS (2001)
    Google Scholar
  10. Wersing, H., Steil, J.J., Ritter, H.: A Competitive Layer Model for Feature Binding and Sensory Segmentation, Neural Computation vol. 13 (2001) 357–387.
    Article MATH Google Scholar

Download references

Author information

Authors and Affiliations

  1. Neuroinformatics Department, P.O.-Box 100131, D-33501, Bielefeld, Germany
    Sebastian Weng & Jochen J. Steil

Authors

  1. Sebastian Weng
  2. Jochen J. Steil

Editor information

Editors and Affiliations

  1. ETS Informática, Universidad Autónoma de Madrid, 28049, Madrid, Spain
    José R. Dorronsoro

Rights and permissions

© 2002 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Weng, S., Steil, J.J. (2002). Data Driven Generation of Interactions for Feature Binding and Relaxation Labeling. In: Dorronsoro, J.R. (eds) Artificial Neural Networks — ICANN 2002. ICANN 2002. Lecture Notes in Computer Science, vol 2415. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-46084-5\_70

Download citation

Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Publish with us