QDS-SNN: Quantum spiking neural network cuts traffic sign recognition energy by 55%

Traffic sign recognition is critical for autonomous driving but traditional deep learning methods are compute- and energy-hungry. Spiking Neural Networks (SNNs) offer a bio-inspired, low-power alternative, but they suffer from information loss and vanishing gradients. A new paper from researchers at multiple institutions introduces QDS-SNN, which fuses quantum neural networks (QNNs) with spiking neurons to deliver both high accuracy and extreme energy efficiency.

The model uses a temporally and spatially adaptive LIF (TSA-LIF) neuron and a quantum-assisted classifier module (QACM) to mitigate gradient issues. Simulated on PennyLane, QDS-SNN achieves 99.72% accuracy on the GTSRB dataset in just 6 time steps—outperforming the MS-ResNet baseline by 1.32% while reducing energy consumption by 55.77%. On the TSRD dataset, accuracy reaches 97.90% with energy use at only 52.68% of baseline. This work demonstrates that quantum-spiking hybrid systems can deliver real-time, low-power inference for intelligent transportation.