DOI QR코드

DOI QR Code

Acceleration Method of Inter Prediction using Advanced SIMD

Advanced SIMD를 이용한 화면 간 예측 고속화방법

  • Kim, Wan-Su (Dept. of Computer Engineering, Hanbat University) ;
  • Lee, Jae-Heung (Dept. of Computer Engineering, Hanbat University)
  • Received : 2012.12.09
  • Accepted : 2012.12.17
  • Published : 2012.12.31

Abstract

An H.264/AVC fast motion estimation methodology is presented in this paper. Advanced SIMD based NEON which is one of the parallel processing methods is supported under the ARM Cortex-A9 dual-core platform. NEON is applied to a full search technique with one of the various motion estimation methods and SAD operation count of each macroblock is reduced to 1/4. Pixel values of the corresponding macroblock are assigned to eight 16-bit NEON registers and Intrinsic function in NEON architecture carried out 128 bits arithmetic operations at the same time. In this way, the exact motion vector with the minimum SAD value among the calculated SAD values can be designated. Experimental results show that performance gets improved 30% above average in accordance with the size of image and macroblock.

본 논문에서는 ARM Cortex-A9 듀얼코어 플랫폼에서 지원하는 병렬처리 기법 중 하나인 Advanced SIMD기반의 NEON을 적용한 H.264/AVC 고속화 움직임추정 방법론을 연구하였다. 다양한 움직임추정 방법 중 하나인 전역탐색기법에 NEON을 적용하여 각 매크로블록의 SAD 연산횟수를 1/4 감소시켰다. 해당 매크로블록의 픽셀 값들을 8개의 16bit NEON 레지스터에 할당하였고, NEON에서 지원하는 Intrinsic 함수를 사용하여 동시에 128bit 연산을 수행하였다. 이러한 방법으로 계산된 SAD 값들 중 최소 SAD 값을 가지고 정확한 모션벡터를 선정했다. 그 결과 영상의 크기 및 매크로블록의 크기에 따라 성능이 평균 30% 이상 향상되는 효과를 검증하였다.

Keywords

References

  1. Jae-Chang Jung, H.264/AVC Video Compress Standard, Hong-Rling Pub, Seoul, pp.2-5, 2005.
  2. ARM, Cortex-A9 MPCore Technical Reference Manual, http://infocenter.arm.com/, 2010.
  3. ARM, Introducing NEON - Development Article, http://infocenter.arm.com/, 2009.
  4. Iain E. G. Richardson, H.264 and MPEG-4 Video Compression: Video Coding for Next generation Multimedia, John Wiley & Sons, Hoboken, pp.32-33, 2003.
  5. Yu-Wen Hung, Bing-Yu Huang ,Tu-Chih Wang, "Analysis and Reduction of Reference Frames for Mo tion Estimation in MPEG-4 AVC/JVT/H.264", Multimedia and Expo, Vol.2, pp.II-809-12, July. 2003.
  6. Chirag Pujara et al, "H.264 Video Decoder Optimization on ARM Cortex-A8 with NEON", INDICON 2009 Anual IEEE, pp.1-3, Dec. 2009.
  7. ARM, RealView Compiler Tools v3.1 - NEON Compiler, http://infocenter.arm.com/, 2007.
  8. Young-Hoon Jung, SIMD Parallel Programming, Freelec, Bucheon, pp.215-544, 2012.
  9. Wing-Yee Lo, Daniel Pak-Kong Lun, Wan-Chi Siu, "Improved SIMD Architecture for High Performan ce Video Processors", Circuits and System for Video Technology, IEEE Transaction on, Vol.21, pp.1769-1783, Dec. 2011. https://doi.org/10.1109/TCSVT.2011.2130250