Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                

CAVIAR: a 45k neuron, 5M synapse, 12G connects/s AER hardware sensory-processing- learning-actuating system for high-speed visual object recognition and tracking

IEEE Trans Neural Netw. 2009 Sep;20(9):1417-38. doi: 10.1109/TNN.2009.2023653. Epub 2009 Jul 24.

Abstract

This paper describes CAVIAR, a massively parallel hardware implementation of a spike-based sensing-processing-learning-actuating system inspired by the physiology of the nervous system. CAVIAR uses the asychronous address-event representation (AER) communication framework and was developed in the context of a European Union funded project. It has four custom mixed-signal AER chips, five custom digital AER interface components, 45k neurons (spiking cells), up to 5M synapses, performs 12G synaptic operations per second, and achieves millisecond object recognition and tracking latencies.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Action Potentials
  • Artificial Intelligence*
  • Computers
  • Humans
  • Learning / physiology
  • Motion Perception / physiology
  • Neural Networks, Computer*
  • Neurons / physiology
  • Pattern Recognition, Visual* / physiology
  • Psychomotor Performance* / physiology
  • Retina / physiology
  • Synapses / physiology
  • Time Factors
  • Vision, Ocular* / physiology
  • Visual Perception* / physiology