Face-recognition hardware implementation based on sope

  • Authors:

  • Cheng He;Chunchun Sui;Jing Chen;Cao Xixin

  • Affiliations:

  • School of Software and Microelectronic, Peking University, Beijing, China;School of Software and Microelectronic, Peking University, Beijing, China;School of Software and Microelectronic, Peking University, Beijing, China;School of Software and Microelectronic, Peking University, Beijing, China

  • Venue:
  • ICACT'10 Proceedings of the 12th international conference on Advanced communication technology
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

As the rapid development of computer science and pattern recognition, visual technology is widely applied in industry and daily life. This paper provides a face-recognition hardware implementation based on SOPC, in which custom peripherals, manufacture-provided IP and NiosII processor are connected by Avalon Memory-Mapped Interface. Real time data is transported through Avalon Streaming Interface between custom peripherals without the intervening of CPU. Two kinds of hardware acceleration are used to improve performance, C2H and Custom instructions, which can accelerate the speed of face-recognition up to about 50 and 90 times, respectively.