arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2311.18655cs.AReess.SP
收藏

OISA:构建一种用于高效视觉计算的光学传感器内加速器架构

OISA: Architecting an Optical In-Sensor Accelerator for Efficient Visual Computing

Mehrdad Morsali, Sepehr Tabrizchi, Deniz Najafi, Mohsen Imani, Mahdi Nikdast, Arman Roohi, Shaahin Angizi

AI总结:

本文首次提出光学传感器内加速器架构 OISA,通过最小转换方式在低比特神经网络中实现粗粒度卷积,在保持可接受精度的同时显著提升能效并降低功耗。

摘要:

面向边缘视觉应用,本文首次系统地探索并提出了一种高性能、高能效的光学传感器内加速器架构,称为 OISA。利用光子器件所具备的良好效率,OISA 在低比特宽度神经网络中以创新的最小转换方式,对输入帧直接实现粗粒度卷积运算。这种设计显著降低了传统以云为中心的架构以及近期提出的边缘加速器中数据转换、传输和处理的功耗。我们在多种图像数据集上进行的从器件到架构的仿真结果表明,OISA 在实现可接受精度的同时,能效达到 6.68 TOp/s/W。与现有的电子传感器内/近传感器加速器和 ASIC 加速器相比,OISA 平均将功耗降低了 7.9 倍和 18.4 倍。

英文摘要:

Targeting vision applications at the edge, in this work, we systematically explore and propose a high-performance and energy-efficient Optical In-Sensor Accelerator architecture called OISA for the first time. Taking advantage of the promising efficiency of photonic devices, the OISA intrinsically implements a coarse-grained convolution operation on the input frames in an innovative minimum-conversion fashion in low-bit-width neural networks. Such a design remarkably reduces the power consumption of data conversion, transmission, and processing in the conventional cloud-centric architecture as well as recently-presented edge accelerators. Our device-to-architecture simulation results on various image data-sets demonstrate acceptable accuracy while OISA achieves 6.68 TOp/s/W efficiency. OISA reduces power consumption by a factor of 7.9 and 18.4 on average compared with existing electronic in-/near-sensor and ASIC accelerators.

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。

↑