arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2404.19489cs.CVcs.ARcs.ETcs.NE
收藏

EvGNN:一种用于边缘视觉的事件驱动图神经网络加速器

EvGNN: An Event-driven Graph Neural Network Accelerator for Edge Vision

Delft University of Technology(代尔夫特理工大学);KU Leuven(鲁汶大学)

Yufeng Yang, Adrian Kneip, Charlotte Frenkel

AI总结:

本文提出首款面向边缘视觉的事件驱动图神经网络加速器EvGNN,通过有向动态图、事件队列和层并行处理实现低延迟高精度推理,在N-CARS数据集上达87.8%精度与每事件16μs延迟。

摘要:

结合传感与嵌入式处理的边缘视觉系统有望提供低延迟、去中心化且高能效的解决方案,从而摒弃对云端的依赖。与传统的基于帧的视觉传感器不同,基于事件的相机以微秒级的时间分辨率和稀疏信息编码方式传递数据,从而为边缘视觉系统勾勒出新的机遇。然而,主流的基于帧的视觉算法大多依赖卷积神经网络(CNNs),由于它们通常针对密集矩阵-向量乘法进行优化,因此很难发挥基于事件视觉的优势。尽管事件驱动的图神经网络(GNNs)近期已成为稀疏事件视觉的一种极具前景的解决方案,但其不规则的结构是一项挑战,目前阻碍了高效硬件加速器的设计。在本文中,我们提出了EvGNN,这是首款面向基于事件相机的低占用、超低延迟和高精度边缘视觉的事件驱动GNN加速器。它依赖三个核心思想:(i) 利用单跳节点和无边存储的有向动态图;(ii) 利用事件队列在时空解耦的搜索范围内高效识别局部邻居;以及(iii) 一种允许多层GNN低延迟执行的新型层并行处理方案。我们将EvGNN部署在Xilinx KV260 Ultrascale+ MPSoC平台上,并在N-CARS数据集上对其进行了汽车识别基准测试,证明了其87.8%的分类精度和每个事件16μs的平均延迟,从而在边缘实现了实时、微秒分辨率的事件视觉。

英文摘要:

Edge vision systems combining sensing and embedded processing promise low-latency, decentralized, and energy-efficient solutions that forgo reliance on the cloud. As opposed to conventional frame-based vision sensors, event-based cameras deliver a microsecond-scale temporal resolution with sparse information encoding, thereby outlining new opportunities for edge vision systems. However, mainstream algorithms for frame-based vision, which mostly rely on convolutional neural networks (CNNs), can hardly exploit the advantages of event-based vision as they are typically optimized for dense matrix-vector multiplications. While event-driven graph neural networks (GNNs) have recently emerged as a promising solution for sparse event-based vision, their irregular structure is a challenge that currently hinders the design of efficient hardware accelerators. In this paper, we propose EvGNN, the first event-driven GNN accelerator for low-footprint, ultra-low-latency, and high-accuracy edge vision with event-based cameras. It relies on three central ideas: (i) directed dynamic graphs exploiting single-hop nodes with edge-free storage, (ii) event queues for the efficient identification of local neighbors within a spatiotemporally decoupled search range, and (iii) a novel layer-parallel processing scheme allowing for a low-latency execution of multi-layer GNNs. We deployed EvGNN on a Xilinx KV260 Ultrascale+ MPSoC platform and benchmarked it on the N-CARS dataset for car recognition, demonstrating a classification accuracy of 87.8% and an average latency per event of 16$μ$s, thereby enabling real-time, microsecond-resolution event-based vision at the edge.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑