首页> 外文会议> >BinaryEye: A 20 kfps Streaming Camera System on FPGA with Real-Time On-Device Image Recognition Using Binary Neural Networks

【24h】

BinaryEye: A 20 kfps Streaming Camera System on FPGA with Real-Time On-Device Image Recognition Using Binary Neural Networks

机译：BinaryEye：基于FPGA的20 kfps流媒体摄像头系统，具有使用二进制神经网络进行实时设备上图像识别的功能

获取原文

页面导航

摘要
著录项
相似文献
相关主题

摘要

Streaming high-speed cameras pose a major challenge to distributed cyber-physical and IoT systems, because large data volumes need to be transferred under stringent realtime constraints. Edge processing can mitigate the data deluge by extracting relevant information from image data on-device with low latency. This work presents an FPGA-based 20 kfps streaming camera system, which can classify regions of interest (ROI) within a frame with a binarized neural network (BNN) in realtime streaming mode, achieving massive data reduction. BNNs have the potential to enable energy-efficient image classifications for on-device processing. We demonstrate our system in a case study with a simple real-time BNN classifier achieving 19.28 us latency at 0.52 W power consumption and resulting in a 980x data reduction. We compare external image processing with this result, showing 3x energy savings, and discuss the used HDL/HLS design flow for BNN implementation.

机译：由于需要在严格的实时约束下传输大数据量，因此流式高速摄像头对分布式网络物理和物联网系统构成了重大挑战。边缘处理可以通过以低延迟从设备上的图像数据中提取相关信息来减轻数据泛滥。这项工作提出了一个基于FPGA的20 kfps流媒体摄像头系统，该系统可以在实时流媒体模式下使用二值神经网络（BNN）对帧内的感兴趣区域（ROI）进行分类，从而实现海量数据缩减。 BNN具有实现高能效图像分类以进行设备上处理的潜力。我们在一个案例研究中演示了我们的系统，该案例使用简单的实时BNN分类器以0.52 W的功耗实现了19.28 us的延迟，并减少了980倍的数据。我们将外部图像处理与该结果进行比较，显示出3倍的节能量，并讨论了用于BNN实施的HDL / HLS设计流程。

著录项

来源
《》|2018年|1-7|共7页
会议地点
作者
Petar Jokic; Stephane Emery; Luca Benini;
展开▼
作者单位

展开▼
会议组织
原文格式 PDF
正文语种
中图分类
关键词
Cameras; Streaming media; Field programmable gate arrays; Image recognition; Real-time systems; Random access memory; Neural networks;

机译：相机;流媒体;现场可编程门阵列;图像识别;实时系统;随机存取存储器;神经网络;

相似文献

外文文献
中文文献
专利

1. Real-Time Underwater Image Recognition with FPGA Embedded System for Convolutional Neural Network [J] . Zhao Minghao, Hu Chengquan, Wei Fenglin, Nature reviews Cancer . 2019,第2期

机译：卷积神经网络FPGA嵌入式系统的实时水下图像识别
2. FPGA-Based Implementation of a Real-Time Object Recognition System Using Convolutional Neural Network [J] . Gilan Ali Azarmi, Emad Mohammad, Alizadeh Bijan Circuits and Systems II: Express Briefs, IEEE Transactions on . 2020,第4期

机译：基于FPGA的使用卷积神经网络实现实时对象识别系统
3. StereoEngine: An FPGA-Based Accelerator for Real-Time High-Quality Stereo Estimation With Binary Neural Network [J] . Chen Gang, Ling Yehua, He Tao, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems . 2020,第11期

机译：StereoEnengine：基于FPGA的加速器，用于使用二元神经网络的实时高质量立体声估算
4. BinaryEye: A 20 kfps Streaming Camera System on FPGA with Real-Time On-Device Image Recognition Using Binary Neural Networks [C] . Petar Jokic, Stephane Emery, Luca Benini IEEE International Symposium on Industrial Embedded Systems . 2018

机译：Binaryeye：使用二进制神经网络的FPGA上的20 kFPS流式摄像机系统，使用二进制神经网络实时拍摄图像识别
5. Real-time three-dimensional head position tracker system with stereo cameras using a face recognition neural network. [D] . Girado, Javier Ignacio. 2004

机译：使用面部识别神经网络的带有立体摄像机的实时三维头部位置跟踪器系统。
6. Real-Time Underwater Image Recognition with FPGA Embedded System for Convolutional Neural Network [O] . Minghao Zhao, Chengquan Hu, Fenglin Wei, 2019

机译：用于卷积神经网络的FPGA嵌入式系统实时水下图像识别
7. FPGA implementation of an image recognition system based on tiny neural networks and on-line reconfiguration [O] . Moreno González Félix Antonio, Alarcón Celis Jaime, Salvador Perea Rubén, 2008

机译：基于微型神经网络和在线重配置的图像识别系统的FPGA实现

BinaryEye: A 20 kfps Streaming Camera System on FPGA with Real-Time On-Device Image Recognition Using Binary Neural Networks

摘要

著录项

相似文献

相关主题

期刊订阅