arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.01226cs.LGcs.AR
收藏

低分辨率红外阵列隐私保护人数统计的DNN软硬件协同优化

HW-SW Optimization of DNNs for Privacy-preserving People Counting on Low-resolution Infrared Arrays

Politecnico di Torino(都灵理工大学);ST Microelectronics S.r.l.(意法半导体公司)

Matteo Risso, Chen Xie, Francesco Daghero, Alessio Burrello, Seyedmorteza Mollaei, Marco Castellano, Enrico Macii, Massimo Poncino, Daniele Jahier Pagliari

AI总结:

针对低分辨率红外阵列的隐私保护人数统计,提出DNN全栈自动化优化流程,实现等精度下模型尺寸、代码和能耗的大幅缩减。

摘要:

低分辨率红外(IR)阵列传感器支持人数统计应用,如监测空间占用率和人流,同时保护隐私并最小化能耗。深度神经网络(DNN)已被证明能够准确且高效地处理这些传感器数据。然而,DNN架构空间巨大,手动探索既繁琐又常导致次优解。为克服此问题,我们提出一种高度自动化的全栈DNN优化流程,涵盖神经架构搜索、混合精度量化、后处理,直至实现包含定制指令集微控制器的新型智能传感器原型。通过整合这些跨层优化,我们在能量、内存和精度三维空间中获得大量Pareto最优解。将这些解部署到我们的硬件平台,我们在等精度下实现了高达4.2倍模型尺寸缩减、23.8倍代码尺寸缩减和15.38倍能量缩减,优于现有技术水平。

英文摘要:

Low-resolution infrared (IR) array sensors enable people counting applications such as monitoring the occupancy of spaces and people flows while preserving privacy and minimizing energy consumption. Deep Neural Networks (DNNs) have been shown to be well-suited to process these sensor data in an accurate and efficient manner. Nevertheless, the space of DNNs' architectures is huge and its manual exploration is burdensome and often leads to sub-optimal solutions. To overcome this problem, in this work, we propose a highly automated full-stack optimization flow for DNNs that goes from neural architecture search, mixed-precision quantization, and post-processing, down to the realization of a new smart sensor prototype, including a Microcontroller with a customized instruction set. Integrating these cross-layer optimizations, we obtain a large set of Pareto-optimal solutions in the 3D-space of energy, memory, and accuracy. Deploying such solutions on our hardware platform, we improve the state-of-the-art achieving up to 4.2x model size reduction, 23.8x code size reduction, and 15.38x energy reduction at iso-accuracy.

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。

↑