DRiVE: Dynamic Recognition in VEhicles using snnTorch

TL;DR

This paper introduces DRiVE, a vehicle detection model using Spiking Neural Networks integrated with snnTorch, achieving high accuracy and demonstrating SNNs' potential for image classification tasks beyond temporal data.

Contribution

The study presents a novel SNN-based vehicle detection model, DRiVE, demonstrating effective image classification and challenging the belief that SNNs are limited to temporal data processing.

Findings

DRiVE achieves 94.8% accuracy in vehicle classification.

DRiVE attains a 0.99 AUC score, indicating excellent discrimination.

Results suggest SNNs can be effective for visual tasks beyond temporal data.

Abstract

Spiking Neural Networks (SNNs) mimic biological brain activity, processing data efficiently through an event-driven design, wherein the neurons activate only when inputs exceed specific thresholds. Their ability to track voltage changes over time via membrane potential dynamics helps retain temporal information. This study combines SNNs with PyTorch's adaptable framework, snnTorch, to test their potential for image-based tasks. We introduce DRiVE, a vehicle detection model that uses spiking neuron dynamics to classify images, achieving 94.8% accuracy and a near-perfect 0.99 AUC score. These results highlight DRiVE's ability to distinguish vehicle classes effectively, challenging the notion that SNNs are limited to temporal data. As interest grows in energy-efficient neural models, DRiVE's success emphasizes the need to refine SNN optimization for visual tasks. This work encourages broader exploration of SNNs in scenarios where conventional networks struggle, particularly for real-world applications requiring both precision and efficiency.