Neuromorphic selective attention systems

Selective attention mechanisms allow sensory systems with limited processing capacity to function in real time independent of the size of the stimulus input space. They can be particularly useful in vision systems, where the amount of information provided by the sensors typically exceeds the system´s processing capacity. We present circuits for implementing models of selective attention mechanisms and show application examples, ranging from single chip focal plane processors to active multi chip vision systems, that perform active tracking and selective scanning tasks.