Abstract
Hierarchical neural networks require the parallel extraction of multiple features. This raises the question how a subpopulation of cells can become specific to one feature and invariant to another, while a different subpopulation becomes invariant to the first but specific to the second feature. Using a colour image sequence recorded by a camera mounted to a cat’s head, we train a population of neurons to achieve optimally stable responses. We find that colour sensitive cells emerge. Adding the additional objective of decorrelating the neurons’ outputs leads a subpopulation to develop achromatic receptive fields. The colour sensitive cells tend to be non-oriented, while the achromatic cells are orientation-tuned, in accordance with physiological findings. The proposed objective thus successfully separates cells which are specific for orientation and invariant to colour from orientation invariant colour cells.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Olshausen, B.A., Field, D.J.: Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 381 (1996) 607–609
Hyvärinen, A., Hoyer, P.O.: Emergence of phase and shift invaraint features by decomposition of natural images into independent feature subspaces. Neural Comput. 12 (2000) 1705–1720
Földiak, P.: Learning Invariance from Transformation Sequences. Neural Computation 3 (1991) 194–200
Hurri, J., Hyvärinen A. Simple-Cell-Like Receptive Fields Maximize Temporal Coherence in Natural Video. submitted (2002)
Kayser, C., Einhäuser, W., Dümmer O., König P., Körding K.P.: Extracting slow subspaces from natural videos leads to complex cells. In G. Dorffner, H. Bischoff, K. Hornik (eds.) Artificial Neural Networks-(ICANN) LNCS 2130, Springer-Verlag, Berlin Heidelberg New York (2001) 1075–1080
Gouras, P.: Opponent-colour cells in different layers of foveal striate cortex. J. Physiol 199 (1974) 533–547
Lennie, P., Krauskopf, J., Sclar, G.: Chromatic Mechanisms in Striate Cortex of Macaque. J. Neurosci. 10 (1990) 649–669
Jähne, B.: Digital Image Processing-Concepts, Algortihms and Scientific Applications, 4th compl. rev. edn. Springer-Verlag, Berlin Heidelberg New York (1997)
Hoyer, P.O., Hyvärinen, A.: Independent Component Analysis Applied to Feature Extraction From Colour and Stereo Images. Network 11 (2000) 191–210
Einhäuser, W., Kayser, C, König, P., Körding, K.P.: Learning of complex cells properties from their responses to natural stimuli. Eur. J. Neurosci. 15 (2002) in press.
Lee, T.W., Wachtler, T., Sejnowski, T.: Color Opponency Constitutes A Sparse Representation For the Chromatic Structure of Natural Scenes. NIPS. 13 (2001) 866–872.
Wachtler, T., Lee, T.W., Sejnowski, T.: Chromatic Structure of natural scenes. J. Opt. Sco. Am. A 18 (2001) 65–77
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Einhäuser, W., Kayser, C., Körding, K.P., König, P. (2002). Learning Multiple Feature Representations from Natural Image Sequences. 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_4
Download citation
DOI: https://doi.org/10.1007/3-540-46084-5_4
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-44074-1
Online ISBN: 978-3-540-46084-8
eBook Packages: Springer Book Archive