← 返回 JSSC 论文列表
📄 下载 JSSC 原文 PDF
JSSC 2019第4期Image Sensors130nmCMOS Image Sensor

A 5500-frames/s 85-GOPS/W 3-D Stacked BSI Vision Chip Based on Parallel In-Focal-Plane Acquisition and Processing Laurent Millet , Stephane Chevobbe, Caaliph Andriamisaina, Lamine Benaissa, Edouard Deschaseaux, Edith Beigne , Karim Ben Chehida, Maria Lepecq, Mehdi Darouich, Fabrice Guellec, Thomas Dombek, and

本文介绍了一种基于3D堆叠技术的视觉芯片,具有高帧率和高能效比。
5500帧/s, 85 GOPS/W
3D堆叠视觉芯片高帧率能效比多流处理
创新点1:3D堆叠架构(系统创新) - 采用三维堆叠技术,将图像传感器置于顶层,处理单元和存储器置于底层,实现了高密度集成和短距离互连,显著提升了数据传输效率和能效比(85 GOPS/W)。
创新点2:焦平面读出与计算架构紧密耦合(方法创新) - 通过焦平面内读出技术与可配置计算架构的直接耦合,实现了低延迟(5500帧/秒)和高并行度的图像处理,适用于高速图像分析场景。
创新点3:多流处理能力(系统创新) - 支持在不同区域执行不同程序的多流处理能力,增强了芯片的灵活性和多功能性,适用于复杂视觉任务的多任务并行处理。
创新点4:八方向像素间通信(电路创新) - 通过实现八方向像素间通信,支持大核卷积操作,提升了图像处理的精度和灵活性,扩展了芯片的应用范围。
Abstract
This paper presents a 3-D stacked vision chip featuring in-focal-plane read-out tightly coupled with flexible computing architecture for configurable high-speed image analy- sis. The chip architecture is based on a scalable standalone structure integrating image sensor on the top tier and processing elements (PEs) plus memories i n the bottom tier. By using 3-D stacking partitioning, our prototype benefits from backside illu- minated pixels sensitivity, a fully parallel communication between image sensor and PEs for low-latency performances, while leaving enough room in the bottom tier to embed advanced computing features. One scalable structure embeds a 16 ×16 pixel array (or 64 × 64 pixels in high-resolution mode), associated with an 8-bit single instruction multiple data (SIMD) processor; fabricated in dual 130-nm 1P6M CMOS process. This paper exhibits a 5500 frames/s and 85 giga operations per second (GOPS)/W in low-resolution mode, with large kernels capabilities through eight directions interpixel communication. Multiflow capability is also demonstrated to execute different programs in different areas of the vision chip.