Integer-Valued Training and Spike-Driven Inference Spiking Neural Network for High-performance and Energy-efficient Object Detection

Xinhao Luo,Man Yao,Yuhong Chou,Bo Xu,Guoqi Li

2024-08-05

Abstract:Brain-inspired Spiking Neural Networks (SNNs) have bio-plausibility and low-power advantages over Artificial Neural Networks (ANNs). Applications of SNNs are currently limited to simple classification tasks because of their poor performance. In this work, we focus on bridging the performance gap between ANNs and SNNs on object detection. Our design revolves around network architecture and spiking neuron. First, the overly complex module design causes spike degradation when the YOLO series is converted to the corresponding spiking version. We design a SpikeYOLO architecture to solve this problem by simplifying the vanilla YOLO and incorporating meta SNN blocks. Second, object detection is more sensitive to quantization errors in the conversion of membrane potentials into binary spikes by spiking neurons. To address this challenge, we design a new spiking neuron that activates Integer values during training while maintaining spike-driven by extending virtual timesteps during inference. The proposed method is validated on both static and neuromorphic object detection datasets. On the static COCO dataset, we obtain 66.2% mAP@50 and 48.9% mAP@50:95, which is +15.0% and +18.7% higher than the prior state-of-the-art SNN, respectively. On the neuromorphic Gen1 dataset, we achieve 67.2% mAP@50, which is +2.5% greater than the ANN with equivalent architecture, and the energy efficiency is improved by 5.7*. Code: <a class="link-external link-https" href="https://github.com/BICLab/SpikeYOLO" rel="external noopener nofollow">this https URL</a>

Artificial Intelligence

What problem does this paper attempt to address?

### Problems the Paper Aims to Solve

This paper aims to address the following issues:

1. **Narrowing the Performance Gap between SNN and ANN in Object Detection**: Although brain-inspired Spiking Neural Networks (SNNs) have the advantages of biological plausibility and low power consumption, their performance is currently limited to simple classification tasks. This paper focuses on narrowing the performance gap between SNNs and Artificial Neural Networks (ANNs) in object detection tasks.

2. **Designing an Object Detection Architecture Suitable for SNNs**: Directly converting YOLO series networks to SNN versions leads to spike degradation. Therefore, this paper designs a simplified version of the YOLO architecture (SpikeYOLO), incorporating meta-SNN modules to address this issue.

3. **Reducing Quantization Error**: To address the sensitivity of SNNs to quantization errors in object detection, this paper proposes a new spiking neuron (I-LIF). During training, integer value activations are used, and during inference, they are converted to binary spikes by extending virtual time steps, thereby reducing quantization errors.

Through the above methods, the paper validates the effectiveness of the proposed methods on the static COCO dataset and the neuromorphic Gen1 dataset, significantly improving the performance of SNNs in object detection tasks while maintaining the advantage of low power consumption.