DOI QR코드

DOI QR Code

An Efficient Search Method for Binary-based Block Motion Estimation

이진 블록 매칭 움직임 예측을 위한 효율적인 탐색 알고리듬

  • Lim, Jin-Ho (Department of Electronics Computer Engineering, Hanyang University) ;
  • Jeong, Je-Chang (Department of Electronics Computer Engineering, Hanyang University)
  • 임진호 (한양대학교 전자컴퓨터통신공학과) ;
  • 정제창 (한양대학교 전자컴퓨터통신공학과)
  • Received : 2011.04.18
  • Accepted : 2011.07.04
  • Published : 2011.07.31

Abstract

Motion estimation using one-bit transform and two-bit transform reduces the complexity for computation of matching error; however, the peak signal-to-noise ratio (PSNR) is degraded. Modified 1BT (M1BT) and modified 2BT (M2BT) have been proposed to compensate degraded PSNR by adding conditional local search. However, these algorithms require many additional search points in fast moving sequences with a block size of $16{\times}16$. This paper provides more efficient search method by preparing candidate blocks using the number of non-matching points (NNMP) than the conditional local search. With this NNMP-based search, we can easily obtain candidate blocks with small NNMP and efficiently search final motion vector. Experimental results show that the proposed algorithm not only reduces computational complexity, but also improves PSNR on average compared with conventional search algorithm used in M1BT, M2BT and AM2BT.

1비트 변환 (one-bit transform) 및 2비트 변환 (two-bit transform)을 이용하는 이진 블록 매칭 움직임 예측 (motion estimation) 방법은 전역 탐색 (full search) 움직임 예측 방법에 비해 블록 매칭 연산의 복잡도를 감소시키지만 PSNR (Peak Signal-to-Noise Ratio)성능 저하를 야기한다. 이러한 이진 블록 매칭 움직임 예측 방법의 정확도를 개선하기 위해 조건부 국부 탐색 (conditional localsearch)이 더해져 보완된 1비트 변환 (modified one-bit transform) 및 보완된 2비트 변환 (modified two-bit transform) 방법이 제안되었다. 그러나 이와 같이 추가적인 국부 탐색은 움직임이 빠른 영상에 대한 $16{\times}16$ 블록 크기의 움직임 예측에 있어서 많은 수의 추가적인 탐색을 필요로 한다. 본 논문은 기존의 조건부 국부 탐색 방법 대신 탐색 영역내의 각 후보 블록들의 (candidate blocks) NNMP(Number of Non-Matching Points)를 기반으로 한 효율적인 탐색 방법을 제안한다. NNMP 기반 탐색 방법을 통하여 작은 NNMP 값을 가지는 후보 블록들을 쉽게 탐색하여 최종 움직임 벡터(motion vector)를 효율적으로 찾을 수 있다. 실험을 통하여 제안하는 알고리듬이 기존의 방법들보다 복잡도 및 정확도 측면에서 좋은 성능을 보여주는 것을 확인하였다.

Keywords

References

  1. B. Natarajan, V. Bhaskaran, and K. Konstantinides, "Low-Complexity Block-based Motion Estimation via One-Bit Transforms," IEEE Trans. Circuits and Systems for Video Tech., vol. 7, no. 5, pp. 702-706, Aug. 1997. https://doi.org/10.1109/76.611181
  2. A. Erturk and S.Erturk, "Two-Bit Transform for Binary Block Motion Estimation," IEEE Trans. Circuits and Systems for Video Tech., vol. 15, no. 7, pp. 938-946, July 2005. https://doi.org/10.1109/TCSVT.2005.848340
  3. P. H. W. Wong and O. C. Au, "Modified one-bit transform for motion estimation," IEEE Trans. Circuits Syst. Video Technol., vol. 9, no. 7, pp. 1020-1024, Oct. 1999. https://doi.org/10.1109/76.795055
  4. B. Demir and S. Erturk, "Block motion estimation using modified two bit transform," Lect. Notes in Computer Science, vol. 4263, pp. 522-531, 2006. https://doi.org/10.1007/11902140_56
  5. B. Demir and S. Erturk, "Block Motion Estimation Using Adaptive Modified Two-Bit Transform," IET Image Process., vol. 1, no. 2, pp. 215-222, June 2007. https://doi.org/10.1049/iet-ipr:20060147
  6. Z. He, C. Tsui, K. Chan, and M. Liou, "Low-power VLSI design for motion estimation using adaptive pixel truncation," IEEE Trans. Circuits and Systems for Video Tech., vol. 10, no. 5, pp. 669-678, Aug. 2000. https://doi.org/10.1109/76.856445
  7. J. Feng, K.-T. Lo, H. Mehrpour, and A. E. Karbowiak, "Adaptive block matching motion estimation algorithm using bit-plane matching," in IEEE Int. Conf. Image Processing, Washington, DC, 1995, pp. 496-499. https://doi.org/10.1109/ICIP.1995.537680
  8. Sarp Erturk, "Multiplication-Free One-Bit Transform for Low-Complexity Block-Based Motion Estimation," IEEE Signal Processing Letters, vol. 14, no. 2, 109-112, Feb. 2007. https://doi.org/10.1109/LSP.2006.882088