1997
@techreport{RaB1997,
vgclass = {report},
vgproject = {nn,invariance},
author = {Rajesh P. N. Rao and Dana H. Ballard},
title = {Localized Receptive Fields May Mediate
Transformation-Invariant Recognition in the Visual Cortex},
number = {97.2},
institution = {Department of Computer Science, University of
Rochester},
address = {Rochester, NY 14627},
month = {May},
year = {1997},
abstract = {Neurons in the visual cortex are known to possess
localized, oriented receptive fields. It has previously been suggested
that these distinctive properties may reflect an efficient image
encoding strategy based on maximizing the sparseness of the
distribution of output neuronal activities or alternatively, extracting
the independent components of natural image ensembles. Here, we show
that a relatively simple neural solution to the problem of
transformation-invariant visual recognition also causes localized,
oriented receptive fields to be learned from natural images. These
receptive fields, which code for various transformations in the image
plane, allow a pair of cooperating neural networks, one estimating
object identity (``what'') and the other estimating object
transformations (``where'') to simultaneously recognize an object and
estimate its pose by jointly maximizing the \emph{a posteriori}
probability of generating the observed visual data. We provide
experimental results demonstrating the ability of these networks to
factor retinal stimuli into object-centred features and
object-invariant transformations. The resulting neuronal architecture
suggests concrete computational roles for the neuroanatomical
connections known to exist between the dorsal and ventral visual
pathways.},
}