Lucas Parra, Sarnoff.  Feb. 15, 2001
<br>
<br>
Title:
<br>
Higher-order Statistical Properties Arising from the Non-stationarity 
of Natural Signals
<br>
<br>
Abstract:
<br>
We present evidence that several higher-order statistical properties
of natural images and signals can be explained by a stochastic model
which simply varies scale of an otherwise stationary Gaussian process.
We discuss two interesting consequences.  The first is that a variety
of natural signals can be related through a common model of
spherically invariant random processes, which have the attractive
property that the joint densities can be constructed from the one
dimensional marginal.  The second is that in some cases the
non-stationarity assumption and only second order methods can be
explicitly exploited to find a linear basis that is equivalent to
independent components obtained with higher-order methods. This is
demonstrated on spectro-temporal components of speech.