Abstract
Visual primitives have long been proposed, under the name of geons or generalized cylinders, as components for object models in the brain and in the computer. Geons constitute a structural level intermediate between features and whole objects, and can form a basis for powerful generalization between different view points. We here present a system for the extraction of geon models from images of composite objects. We use synthetic gray-level images, each composed of several shape primitives. Our system finds geons as partial matches between several images. Images and models are represented as arrays of complex cells of Gabor type.
Preview
Unable to display preview. Download preview PDF.
References
G. J. Agin, T. O. Binford, Computer Description of Curved Objects, IJCAI (1973), vol. 3, pp. 629–640.
I. Biederman, Psychological Review 94, 115–147 (1987).
R. Nevatia, G. Medioni, International Journal of Computer Vision 20, 1–42 (1996).
M. Zerroug, R. Nevatia, International Journal of Computer Vision 20, 11–42 (1996).
R. Bajcsy, F. Solina, IEEE Transactions, 231–240 (1987).
S. J. Dickinson, A. P. Pentland, A. Rosenfeld, IEEE Transactions on Pattern Analysis and Machine Intelligence 14, 169–193 (1992).
L. Shams, J. Fiser, A Model for Development of Cortical Lateral Connectivities Using Motion Information (1997)
H. A. Ruff, C. J. Kohler, D. L. Haupt, Developmental Psychology 12, 455–459 (1976).
E. S. Spelke, K. Breinlinger, K. Jacobson, A. Phillips, Perception 22, 1483–1501 (1993).
P. J. Kellman, in Visual Perception and Cognition G. D. IN., Eds. Hillsdale (1989), vol. 1, pp. 121–173.
P. J. Kellman, E. S. Spelke, K. R. Short,Child Development 57, 72–86 (1986).
L. Hainline, P. Riddell, J. Grose-Fiser, I. Abramov, Special Issue: Normal and abnormal visual development in infants and children. Behavioral Brain Research 49, 33–50 (1992).
R. Held, E. E. Birch, J. Gwiazda,Stereoacuity of human infants, Natioanl Academy of Sciences USA (1980), vol. 77
L. Wiskott, IR-INI 96-10, Institut fur Neuroinformatik, Ruhr-Universitat Bochum, 44780 Bochum, Germany (1996).
O. Groetenherdt, Master Thesis, Ruhr University, Bochum (1997).
A. Burkhalter, K. L. Bernardo, V. Charles, The Journal Of Neuroscience 13, 1916–1931 (1993).
M. Lades, et al., IEEE Transaction on Computers 42, 300–311 (1993).
L. Wiskott, C. von der Malsburg, Int. J. Pattern Recognition and Artificial Intelligence 7, 935–948 (1993).
L. Wiskott, in Labeled Graphs and Dynamic Link Matching for Face Recognition and Scene Analysis Frankfurt (1996), vol. Band 53, pp. 73–80.
C. von der Malsburg, in The Neural and Molecular Bases of Learning J.-P. Changeux, M. Konishi, Eds. (John Wiley & Sons Limited, 1987) pp. 411–432.
J. E. Hummel, I. Biederman, Psychological Review 99, 480–517 (1992).
T. V. P. Bliss, T. Lomo, Journal of Physiology 232, 331–356 (1973).
A. H. Klopf, The Hedonistic Neuron: A Theory of Memory, Learning, and Intelligence (Hemisphere, Washington, DC, 1982).
P. Foldiak, Neural Computataion 3, 194–200 (1991).
L. Wiskott, J.-M. Fellous, N. Krueger, C. von der Malsburg, Face Recognition and gender determination, International Workshop on Automatic Face-and Gesture-Recognition Zurich (1995).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Shams, L., von der Malsburg, C. (1997). Development of shape primitives from images of composite objects represented by complex cells. In: Gerstner, W., Germond, A., Hasler, M., Nicoud, JD. (eds) Artificial Neural Networks — ICANN'97. ICANN 1997. Lecture Notes in Computer Science, vol 1327. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0020266
Download citation
DOI: https://doi.org/10.1007/BFb0020266
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63631-1
Online ISBN: 978-3-540-69620-9
eBook Packages: Springer Book Archive