Saliency extraction with a distributed spiking neural network

Sylvain Chevallier (1, 2) , Philippe Tarroux (1, 3) , Hélène Paugam-Moisy (4)

1 LIMSI - Laboratoire d'Informatique pour la Mécanique et les Sciences de l'Ingénieur
2 UP11 - Université Paris-Sud - Paris 11
3 ENS-PSL - École normale supérieure - Paris
4 ISC - Institut des Sciences Cognitives

We present a distributed spiking neuron network (SNN) for handling low-level visual perception in order to extract salient locations in robot camera images. We describe a new method which reduce the computional load of the whole system, stemming from our choices of architecture. We also describe a modeling of post-synaptic potential, which allows to quickly compute the contribution of a sum of incoming spikes to a neuron's membrane potential. The interests of this saliency extraction method, which differs from classical image processing, are also exposed.

Mots clés: spiking neurons; saliency; computer vision; rank order coding

Domaines: Informatique; Neurosciences; Réseau de neurones [cs.NE]; Synthèse d'image et réalité virtuelle [cs.GR]

Sylvain Chevallier, Philippe Tarroux, Hélène Paugam-Moisy. Saliency extraction with a distributed spiking neural network. European Symposium on Artificial Neural Networks, Apr 2006, Bruges, Belgium. pp.209-214. ⟨hal-00023820v2⟩