Real-Time Frame- and Event-based Object Detection with Spiking Neural Networks on Edge Neuromorphic Hardware: Design, Deployment and Benchmark

2026-04-30

AI Summary

This work proposes an end-to-end spiking neural network (SNN) object detection architecture tailored for the Intel Loihi 2 neuromorphic chip, addressing the demand for real-time, low-power inference on edge devices. The framework supports dual-modal inputs—conventional frames and event-based data—and integrates ANN-to-SNN conversion with a distillation-aware training strategy to substantially reduce inference energy while effectively recovering accuracy. Experimental results demonstrate that the proposed method achieves the lowest per-inference dynamic power consumption on Loihi 2, outperforms conventional artificial neural networks in latency, and recovers 87%–100% of the original detection accuracy, thereby validating the energy efficiency and practicality of neuromorphic computing for real-time edge vision tasks.

Abstract

Real-time object detection on energy-constrained platforms is critical for applications such as UAV-based inspection, autonomous navigation, and mobile robotics. Spiking neural networks (SNNs) on neuromorphic hardware are believed to be significantly more energy-efficient than conventional artificial neural networks (ANNs). In this work, we present a comprehensive methodology for designing general SNN detection architectures targeting neuromorphic platforms, along with the engineering adaptations required to deploy them on the state-of-the-art Neuromorphic processor, Intel Loihi 2. We benchmark SNN-based object detection on Loihi 2 using both frame-based and event-based datasets, comparing performance with ANN-based detection on the NVIDIA Jetson Orin Nano, NVIDIA Jetson Nano B01, and the Apple M2 CPU. Our results show that SNNs on Loihi 2 can perform real-time detection while achieving the lowest per-inference dynamic energy among all platforms. Also, Loihi 2 outperforms the other platforms in terms of power consumption, though ANNs on Jetson Orin Nano achieve higher inference rates. Furthermore, our ANN-to-SNN distillation-aware training enables SNNs to recover 87-100% of the detection accuracy of their ANN counterparts while maintaining lower inference latency; without distillation, SNNs exhibit an 11-27% accuracy drop. These results highlight the potential of neuromorphic systems for energy-efficient, real-time object detection at the edge.

Problem

• real-time object detection
• energy-constrained platforms
• spiking neural networks
• neuromorphic hardware
• edge computing

Innovation

• Spiking Neural Networks
• Neuromorphic Hardware
• Object Detection
• ANN-to-SNN Distillation
• Energy Efficiency