Shrinking Your TimeStep: Towards Low-Latency Neuromorphic Object Recognition with Spiking Neural Network

2 Jan 2024  ·  Yongqi Ding, Lin Zuo, Mengmeng Jing, Pei He, Yongjun Xiao ·

Neuromorphic object recognition with spiking neural networks (SNNs) is the cornerstone of low-power neuromorphic computing. However, existing SNNs suffer from significant latency, utilizing 10 to 40 timesteps or more, to recognize neuromorphic objects. At low latencies, the performance of existing SNNs is drastically degraded. In this work, we propose the Shrinking SNN (SSNN) to achieve low-latency neuromorphic object recognition without reducing performance. Concretely, we alleviate the temporal redundancy in SNNs by dividing SNNs into multiple stages with progressively shrinking timesteps, which significantly reduces the inference latency. During timestep shrinkage, the temporal transformer smoothly transforms the temporal scale and preserves the information maximally. Moreover, we add multiple early classifiers to the SNN during training to mitigate the mismatch between the surrogate gradient and the true gradient, as well as the gradient vanishing/exploding, thus eliminating the performance degradation at low latency. Extensive experiments on neuromorphic datasets, CIFAR10-DVS, N-Caltech101, and DVS-Gesture have revealed that SSNN is able to improve the baseline accuracy by 6.55% ~ 21.41%. With only 5 average timesteps and without any data augmentation, SSNN is able to achieve an accuracy of 73.63% on CIFAR10-DVS. This work presents a heterogeneous temporal scale SNN and provides valuable insights into the development of high-performance, low-latency SNNs.

PDF Abstract

Results from the Paper

 Ranked #1 on Event data classification on N-Caltech 101 (Accuracy (% ) metric)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Object Recognition CIFAR10-DVS SSNN Accuracy (% ) 78.57 # 1
Event data classification DVS128 Gesture SSNN Accuracy (% ) 94.91 # 1
Object Recognition DVS128 Gesture SSNN Accuracy (% ) 94.91 # 1
Event data classification N-Caltech 101 SSNN Accuracy (% ) 79.25 # 1
Object Recognition N-Caltech 101 SSNN Accuracy (% ) 79.25 # 1