Leveraging Asynchronous Spiking Neural Networks for Ultra Efficient Event-Based Visual Processing
DingYi Zeng, Yuchen Wang, Honglin Cao, Wanlong Liu, Yichen Xiao, ChengzhuoLu, Wenyu Chen, Malu Zhang
Abstract
Event cameras encode visual information by generating asynchronous and sparse event streams, which hold great potential for low latency and low power consumption. Despite many successful implementations of event camera-based applications, most of them accumulate the events into frames and then utilize conventional frame-based computer vision algorithms. These frame-based methods, though typically effective, diminish the inherent advantages of the event camera's low latency and low power consumption. To solve the above problems, we propose ASGCN, which efficiently processes data on an event-by-event basis and dynamically evolves into a corresponding dynamic representation, enabling low latency and high sparsity of data representation. The sparsity computation is further improved by introducing brain-inspired spiking neural networks, resulting in low power consumption for ASGCN. Extensive and diverse experiments demonstrate the energy efficiency and low latency advantages of our processing pipeline. Especially on real-world event camera datasets, our pipeline consumes more than 10,000 times less energy and achieves similar performance compared to current frame-based methods.
BibTeX
@article{Zeng_Wang_Cao_Liu_Xiao_Chen_Zhang_Wang_Yang_2025, title={Leveraging Asynchronous Spiking Neural Networks for Ultra Efficient Event-Based Visual Processing}, volume={39}, url={https://ojs.aaai.org/index.php/AAAI/article/view/32154}, DOI={10.1609/aaai.v39i2.32154}, abstractNote={Event cameras encode visual information by generating asynchronous and sparse event streams, which hold great potential for low latency and low power consumption. Despite many successful implementations of event camera-based applications, most of them accumulate the events into frames and then utilize conventional frame-based computer vision algorithms. These frame-based methods, though typically effective, diminish the inherent advantages of the event camera’s low latency and low power consumption. To solve the above problems, we propose ASGCN, which efficiently processes data on an event-by-event basis and dynamically evolves into a corresponding dynamic representation, enabling low latency and high sparsity of data representation. The sparsity computation is further improved by introducing brain-inspired spiking neural networks, resulting in low power consumption for ASGCN. Extensive and diverse experiments demonstrate the energy efficiency and low latency advantages of our processing pipeline. Especially on real-world event camera datasets, our pipeline consumes more than 10,000 times less energy and achieves similar performance compared to current frame-based methods.}, number={2}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Zeng, DingYi and Wang, Yuchen and Cao, Honglin and Liu, Wanlong and Xiao, Yichen and , ChengzhuoLu and Chen, Wenyu and Zhang, Malu and Wang, Guoqing and Yang, Yang}, year={2025}, month={Apr.}, pages={1620-1628} }