arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2202.09035cs.AR
收藏

PISA:一种用于边缘图像处理的二值加权传感器内处理加速器

PISA: A Binary-Weight Processing-In-Sensor Accelerator for Edge Image Processing

Shaahin Angizi, Sepehr Tabrizchi, Arman Roohi

摘要:

这项工作提出了一种传感器内处理加速器,即PISA,作为AI设备中实时和智能图像处理的灵活,节能和高性能解决方案。PISA本质上实现了二值化权重神经网络(BWNN)中的粗粒度卷积运算,利用了传感器侧具有非易失性权重存储的新型计算像素。这显著降低了数据转换和传输到片外处理器的功耗。该设计完成了一个逐位的近传感器处理在DRAM计算单元来处理其余的网络层。一旦检测到物体,PISA就会切换到典型的传感模式,仅使用近传感器处理单元捕获图像进行细粒度卷积。我们对BWNN加速的电路到应用程序联合仿真结果表明,与基线BWNN模型相比,在粗粒度评估中,各种图像数据集的准确性均可接受,而PISA实现了1000的帧速率和约1.74 TOp/s/W的效率。最后,与基准CPU传感器设计相比,PISA将数据转换和传输能量大幅降低了约84%。

英文摘要:

This work proposes a Processing-In-Sensor Accelerator, namely PISA, as a flexible, energy-efficient, and high-performance solution for real-time and smart image processing in AI devices. PISA intrinsically implements a coarse-grained convolution operation in Binarized-Weight Neural Networks (BWNNs) leveraging a novel compute-pixel with non-volatile weight storage at the sensor side. This remarkably reduces the power consumption of data conversion and transmission to an off-chip processor. The design is completed with a bit-wise near-sensor processing-in-DRAM computing unit to process the remaining network layers. Once the object is detected, PISA switches to typical sensing mode to capture the image for a fine-grained convolution using only the near-sensor processing unit. Our circuit-to-application co-simulation results on a BWNN acceleration demonstrate acceptable accuracy on various image datasets in coarse-grained evaluation compared to baseline BWNN models, while PISA achieves a frame rate of 1000 and efficiency of ~1.74 TOp/s/W. Lastly, PISA substantially reduces data conversion and transmission energy by ~84% compared to a baseline CPU-sensor design.

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。

↑