Real-time object based single-stream to multi-stream network enabled multimedia system using an adder-less reconfigurable fast area correlator processor

被引:0
|
作者
Khan, FN [1 ]
Khan, SA [1 ]
机构
[1] Fatima Jinnah Women Univ, Rawalpindi, Pakistan
关键词
object detection; real-time system; runtime recofigurable hardware; hig resolution; area correlation; object based single video stream to multi video stream generation;
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper presents a network enabled reconfigurable FPGA based real-time system for selectively detecting, adding, deleting or displacing irregular shaped video objects in high-resolution video sequence obtained from a digital camera. Real-time irregular shaped object detection and insertion from a high-resolution high frame rate video requires a high performance system. The system can selectively add or delete video objects from a single video stream and produce multiple streams for broadcasting in a network. Each host in the network get a different version of the video. In this paper we present a run-time reconfigurable design of an area correlator, which is a hardware efficient technique for detecting objects in a frame. The main contribution resides in the concurrent apporach which modfies the algorithm for hardware specific implementation, novel hardware building blocks, and a set of architectures which can be mapped as reconfigurable design at runtime. This system provides flexibility in creative video enriched multimedia applications and has wide range of applications and is applicable in object detection, video recording, surveillance, conferencing, broadcasting and augmented reality.
引用
收藏
页码:688 / 693
页数:6
相关论文
共 3 条
  • [1] Multi-Stream Siamese and Faster Region-Based Neural Network for Real-Time Object Tracking
    Liu, Yi
    Zhang, Liming
    Chen, Zhihui
    Yan, Yan
    Wang, Hanzi
    IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2021, 22 (11) : 7279 - 7292
  • [2] Real-Time Prediction System of Train Carriage Load Based on Multi-Stream Fuzzy Learning
    Yu, Hang
    Lu, Jie
    Liu, Anjin
    Wang, Bin
    Li, Ruimin
    Zhang, Guangquan
    IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2022, 23 (09) : 15155 - 15165
  • [3] Real-time Speaker Recognition System using Multi-stream i-vectors for AI Assistant
    Cho, Keunseok
    Roh, Jaeyoung
    Han, Youngho
    Kim, Namhoon
    Lee, Jaewon
    2018 IEEE INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS (ICCE), 2018,