Spiking neural networks for frame-based and event-based single object localization

Sami Barchid, José Mennesson, Jason Eshraghian, Chaabane Djéraba, Mohammed Bennamoun

Abstract: Highlights•We propose a convolutional spiking neural network for single object localization.•Our spiking model achieves competitive performance and robustness compared to an ANN.•We study temporal latency’s impact, correlating low latency with better performance.•Our analysis of neural coding shows rate coding as the best trade-off in our context.

External IDs:doi:10.1016/j.neucom.2023.126805