Papers on natural image statistics and V1

[Aapo Hyvarinen]

Dear Colleagues,

the following papers are now available on the web.

------------------------------------------------------------

P.O. Hoyer and A. Hyvarinen.
A non-negative sparse coding network learns contour coding and
integration from natural images. 
Submitted manuscript.
http://www.cis.hut.fi/aapo/ps/gz/contours.ps.gz

Abstract: An important approach in visual neuroscience considers how
the function of the early visual system relates to the statistics of
its natural input.  Previous studies have shown how many basic
properties of the primary visual cortex, such as the receptive fields
of simple and complex cells and the spatial organization (topography)
of the cells, can be understood as efficient coding of natural
images. Here we extend the framework by considering how the responses
of complex cells could be efficiently coded by a higher-order neural
layer. This leads to contour coding and end-stopped receptive
fields. Interestingly, contour integration can in this framework be
seen as a direct result of top-down noise reduction, suggesting such a
role for cortico-cortical feedback connections in the visual cortex.

------------------------------------------------------------

A. Hyvarinen and P.O. Hoyer.
A two-layer sparse coding model learns simple and complex
cell receptive fields and topography from natural images.
Vision Research, in press.
http://www.cis.hut.fi/aapo/ps/gz/VR01.ps.gz

Abstract: The classical receptive fields of simple cells in the visual
cortex have been shown to emerge from the statistical properties of
natural images by forcing the cell responses to be maximally sparse,
i.e. significantly activated only rarely.  Here, we show that this
single principle of sparseness can also lead to emergence of
topography (columnar organization) and complex cell properties as
well.  These are obtained by maximizing the sparsenesses of locally
pooled energies, which correspond to complex cell outputs.  Thus we
obtain a highly parsimonious model of how these properties of the
visual cortex are adapted to the characteristics of the natural input.

------------------------------------------------------------

A. Hyvarinen and M. Inki.
Estimating overcomplete independent component bases for image windows.
Submitted manuscript.
http://www.cis.hut.fi/aapo/ps/gz/JMIV02.ps.gz

Abstract: Estimating overcomplete ICA bases for image windows is a
difficult problem. Most algorithms are based on approximations of the
likelihood, which leads to computationally heavy procedures. Here we
first review the existing methods, and then introduce two algorithms
that are based on heuristic approximations and estimate an approximate
overcomplete basis quite fast. The algorithms are based on
quasi-orthogonality in high-dimensional spaces, and the
gaussianization procedure, respectively. 




----------------------------------------------------
Aapo Hyvarinen

Neural Networks Research Centre
Helsinki University of Technology
P.O.Box 5400, FIN-02015 HUT, Finland
Tel: +358-9-4513278, Fax: +358-9-4513277

Email: Aapo.Hyvarinen at hut.fi
Home page: http://www.cis.hut.fi/aapo/
----------------------------------------------------