An Efficient Real-Time Object Detection Framework on Resource-Constricted Hardware Devices via Software and Hardware Co-design (Invited Paper)

被引:6
|
作者
Liu, Mingshuo [1 ]
Luo, Shiyi [1 ]
Han, Kevin [1 ]
Yuan, Bo [2 ]
DeMara, Ronald F. [3 ]
Bai, Yu [1 ]
机构
[1] Calif State Univ Fullerton, Fullerton, CA 92634 USA
[2] Rutgers State Univ, New Brunswick, NJ USA
[3] Univ Cent Florida, Orlando, FL 32816 USA
来源
2021 IEEE 32ND INTERNATIONAL CONFERENCE ON APPLICATION-SPECIFIC SYSTEMS, ARCHITECTURES AND PROCESSORS (ASAP 2021) | 2021年
基金
美国国家科学基金会;
关键词
Deep learning; FPGA; Tensor Train; Energy Efficiency; RECURRENT NEURAL-NETWORKS;
D O I
10.1109/ASAP52443.2021.00020
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
The fast development of object detection techniques has attracted attention to developing efficient Deep Neural Networks (DNNs). However, the current state-of-the-art DNN models can not provide a balanced solution among accuracy, speed, and model size. This paper proposes an efficient real-time object detection framework on resource-constricted hardware devices through hardware and software co-design. The Tensor Train (TT) decomposition is proposed for compressing the YOLOv5 model. By unitizing the unique characteristics given by the TT decomposition, we develop an efficient hardware accelerator based on FPGA devices. Experimental results show that the proposed method can significantly reduce the model size and improve the execution time.
引用
收藏
页码:77 / 84
页数:8
相关论文
共 50 条
  • [31] HSCoNAS: Hardware-Software Co-Design of Efficient DNNs via Neural Architecture Search
    Luo, Xiangzhong
    Liu, Di
    Huai, Shuo
    Liu, Weichen
    PROCEEDINGS OF THE 2021 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION (DATE 2021), 2021, : 418 - 421
  • [32] Hardware/Software Codesign of Real-Time Intrusion Detection System for Internet of Things Devices
    Zeng, Qingyu
    Hara-Azumi, Yuko
    IEEE INTERNET OF THINGS JOURNAL, 2024, 11 (12): : 22351 - 22363
  • [33] I/O-GUARD: Hardware/Software Co-Design for I/O Virtualization with Guaranteed Real-time Performance
    Jiang, Zhe
    Yang, Kecheng
    Ma, Yunfeng
    Fisher, Nathan
    Audsley, Neil
    Dong, Zheng
    2021 58TH ACM/IEEE DESIGN AUTOMATION CONFERENCE (DAC), 2021, : 1159 - 1164
  • [34] Hardware-Software Co-Design for Real-Time Latency-Accuracy Navigation in Tiny Machine Learning Applications
    Behnam, Payman
    Tong, Jianming
    Khare, Alind
    Chen, Yangyu
    Pan, Yue
    Gadikar, Pranav
    Bambhaniya, Abhimanyu
    Krishna, Tushar
    Tumanov, Alexey
    IEEE MICRO, 2023, 43 (06) : 93 - 101
  • [35] Hardware-Software Co-design of 3D Data Compression for Real-time Ultrasonic Imaging Applications
    Govindan, Pramod
    Saniie, Jafar
    2014 IEEE INTERNATIONAL ULTRASONICS SYMPOSIUM (IUS), 2014, : 564 - 567
  • [36] CPU model-based mechatronics/hardware/software co-design technology for real-time embedded control systems
    Ishikawa, Makoto
    Saikalis, George
    Oho, Shigeru
    IEICE TRANSACTIONS ON ELECTRONICS, 2007, E90C (10) : 1992 - 2001
  • [37] Trading Utilization for Circuitry: Hardware-Software Co-design for Real-Time Software-Based Short-Circuit Protection
    Willcock, Aaron
    Fisher, Nathan
    2017 IEEE 23RD INTERNATIONAL CONFERENCE ON EMBEDDED AND REAL-TIME COMPUTING SYSTEMS AND APPLICATIONS (RTCSA), 2017,
  • [38] Hardware-Software Partitioning for Real-Time Object Detection Using Dynamic Parameter Optimization
    Zaharia, Corneliu
    Popescu, Vlad
    Sandu, Florin
    SENSORS, 2023, 23 (10)
  • [39] IMG-SMP: Algorithm and Hardware Co-Design for Real-time Energy-efficient Neural Motion Planning
    Huang, Lingyi
    Zang, Xiao
    Gong, Yu
    Deng, Chunhua
    Yi, Jingang
    Yuan, Bo
    PROCEEDINGS OF THE 32ND GREAT LAKES SYMPOSIUM ON VLSI 2022, GLSVLSI 2022, 2022, : 373 - 377
  • [40] A Software-Hardware Mixed Design for the FPGA Implementation of the Real-Time Edge Detection
    El Houari, Kobzili
    Cherrad, Benbouchama
    Zohir, Irki
    IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN AND CYBERNETICS (SMC 2010), 2010,