ART-EMAP - A NEURAL-NETWORK ARCHITECTURE FOR OBJECT RECOGNITION BY EVIDENCE ACCUMULATION

A new neural network architecture is introduced for the recognition of pattern classes after supervised and unsupervised learning. Applicati ons include spatio-temporal image understanding and prediction and thr ee dimensional (3-D) object recognition from a series of ambiguous two dimensional views, The architecture, called ART-EMAP, achieves a synt hesis of adaptive resonance theory (ART) and-spatial and temporal evid ence integration for dynamic predictive mapping (EMAP), ART-EMAP exten ds the capabilities of fuzzy ARTMAP in four incremental stages, Stage 1 introduces distributed pattern representation at a view category fie ld, Stage 2 adds a decision criterion to the mapping between view and object categories, delaying identification of ambiguous objects when f aced with a low confidence prediction. Stage 3 augments the system wit h a field where evidence accumulates in medium-term memory, Stage 4 ad ds an unsupervised learning process to fine-tune performance after the limited initial period of supervised network training, Each ART-EMAP stage is illustrated with a benchmark simulation example, using both n oisy and noise-free data. A concluding set of simulations demonstrate ART-EMAP performance on a difficult 3-D object recognition problem.