DOI QR코드

DOI QR Code

Hardware Design of VLIW coprocessor for Computer Vision Application

컴퓨터 비전 응용을 위한 VLIW 보조프로세서의 하드웨어 설계

  • Received : 2014.08.04
  • Accepted : 2014.09.05
  • Published : 2014.09.30

Abstract

In this paper, a VLIW(Very Long Instruction Word) vision coprocessor which can efficiently accelerate computer vision algorithm for automotive is designed. The VLIW coprocessor executes four instructions per clock cycle via 8-stage pipelined structure and has 36 integer and floating-point instructions to accelerate computer vision algorithm for pedestrian detection. The processor has about 300-MHz operating frequency and about 210,900 gates under 45nm CMOS technology and its estimated performance is 1.2 GOPS(Giga Operations Per Second). The vision system composed of vision primitive engine and eight VLIW coprocessors can execute pedestrian detection at 25~29 frames per second(FPS). Because the VLIW coprocessor has high detection rate and loosely coupled interface with host processor, it can be efficiently applicable to a wide range of vision applications.

본 논문에서는 자동차용 컴퓨터 비전 알고리즘을 고속으로 처리하기 위해 VLIW 보조프로세서를 설계하였다. VLIW 보조프로세서는 8단 파이프라인 구조로 1개의 사이클에 4개의 명령을 처리할 수 있으며, 보행자 인식을 위한 36개의 정수 및 부동 소수점 명령어 집합을 갖고 있다. 프로세서는 45nm CMOS 공정에서 최대 동작 속도는 300-MHz이며 약 210,900 게이트로 구성되며 예상 연산 성능은 1.2 GOPS 이다. VPE와 8개의 VLIW 코어로 구성된 비전 프로세서 시스템은 25~29 FPS의 보행자 검출 성능을 가진다. VLIW 보조 프로세서는 높은 검출 속도와 호스트 프로세서와 느슨한 결합 특성으로 다양한 비전 분야에 응용 가능하다.

Keywords

References

  1. Gideon P. Stein, Elchanan Rushineek, Gaby Hayun, and Amnon Shashua, "A Computer Vision System on a Chip : a case study from the automotive domain", Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) Workshops, Volume 03, pp. 130-134. 2005.
  2. Jeff Bier, "Computer Vision in Mobile Devices : The Next Killer App ?", Linley Tech Mobile Device Conference, 2011.
  3. S.Kyo and S. Okazaki, "IMAPCAR: A 100 GOPS in-vehicle vision based 128 ring connected VLIW processing elements," Journal of Signal Processing Systems, vol.62, pp.5-16, 2011. https://doi.org/10.1007/s11265-008-0297-0
  4. Winfried Gerke, Jorn Jachalsky, Martin Wahle, "Flexible Coprocessor Architectures for Ambient Intelligent Applications in Mobile Communication and Automotive Domain," Proc. of SPIE, vol.5117, pp.310-320, 2003.
  5. MobilEye Inc. EyeQ2TM Vision System on a Chip [Internet]. Available: http://www.mobileye.com.
  6. Joseph A Fisher, Paolo Faraboschi, and Cliff Young, Embedded Computing: A VLIW Approach to Architecture, Compilers, and Tools, MK, 2005.
  7. QualComm, "Hexagon V2 Programmer's Reference Manual", 80-NB419-1 Rev.A, August 4, 2011.
  8. Chang-Hyo Yu, Kyusik Chung, Donghyun Kim, Lee-Sup Kim, "An Energy-Efficient Mobile Vertex Processor With Multithread Expanded VLIW Architecture and Vertex Caches", IEEE Journal of Solid State Circuits, vol. 42, no. 10, pp.2256-2269. Oct., 2007.
  9. Mark Zwolinski, Digital System Design with SystemVerilog, Prentice-Hall, 2009.
  10. M. Lam, "Software pipelining: An effective scheduling technique for VLIW machines", PLDI 88, pp. 318-328, July 1988.