How does the nervous system achieve the fundamental task of scene
segmentation? Both theoretical and empirical investigations
point to the mechanism of oscillatory correlation as a potential
representational framework. In this talk, I will describe LEGION (locally
excitatory globally inhibitory oscillator networks), which can
rapidly achieve both synchronization within an assembly of oscillators and
desynchronization between different assemblies. Analytical work on
LEGION has led to an oscillatory correlation approach to scene segmentation
and object selection, and the approach has been applied to
a variety of imagery with considerable success. Efforts have also
been made in VLSI implementation of LEGION networks. The talk will discuss
oscillatory correlation as a neurocomputational framework for addressing
the image analysis problem.
Copyright. All Rights Reserved.