DISTORTION INVARIANT OBJECT RECOGNITION IN THE DYNAMIC LINK ARCHITECTURE

We present an object recognition system based on the Dynamic Link Arch itecture, which is an extension to classical Artificial Neural Network s. The Dynamic Link Architecture exploits correlations in the fine-sca le temporal structure of cellular signals in order to group neurons dy namically into higher-order entities. These entities represent a very rich structure and can code for high level objects. In order to demons trate the capabilities of the Dynamic Link Architecture we implemented a program that can recognize human faces and other objects from video images. Memorized objects are represented by sparse graphs, whose ver tices are labeled by a multi-resolution description in terms of a loca l power spectrum, and whose edges are labeled by geometrical distance vectors. Object recognition can be formulated as elastic graph matchin g, which is performed here by stochastic optimization of a matching co st function. Our implementation on a transputer network successfully a chieves recognition of human faces and office objects from gray level camera images. The performance of the program is evaluated by a statis tical analysis of recognition results from a portrait gallery comprisi ng images of 87 persons.