A Study on the Multiresolution Motion Estimation Adequate to Low-Band-Shift Method in Wavelet Domain

웨이블릿 변환 영역에서 저대역 이동법에 적합한 다해상도 움직임 추정에 관한 연구

  • 조재만 ((주)인텔링스) ;
  • 김현민 ((주)인텔링스) ;
  • 고형화 (광운대학교 전자통신공학과 영상처리연구실)
  • Published : 2003.02.01

Abstract

In this paper, we propose a Multiresolution Motion Estimation(MRME) adapted to Low-Band-Shift(LBS) method in wavelet domain. To overcome shift-variant property on wavelet coefficients, the LBS was previously proposed. This method which is applied to reference frame in video coding technique, has superior performance in terms of rate-distortion characteristic. However, this method needs more memory and computational complexity. In this paper, The computational complexity of the proposed method(LBS-MRME) is about 15.6% of that of existing method at 3-level wavelet transform. And although it has about 7 times as much as existing method's motion vector since each subband has different motion vector, it decreases motion compensated prediction error by detailed motion estimation, and then has better efficient coding performance. The experimental results with the proposed method showed about 0.3∼11.6% improvement of MAD performance in case of lossless coding, and 0.3∼3.0㏈ improvement of PSNR performance at the same bit rate in case of lossy coding.

본 논문에서는 웨이블릿 변환 영역에서 저대역 이동법에 적합한 다해상도 움직임 추정을 제안하였다. 저대역 이동법(Low Band Shift Method)은 웨이블릿 계수들의 이동-변환 성질을 극복하기 위하여 제안된 방법으로 동영상 부호화시 참조 프레임에 적용하면 정확한 움직임 추정이 가능하여 일반적인 방법보다 압축대비 화질면에서 우수한 성능을 가지지만, 단점으로 메모리와 계산량이 일반적인 방법에 비해 많아지게 된다. 본 논문에서 제안된 방법(LBS-MRME)은 저대역 이동법에 적합한 다해상도 움직임 추정을 적용하여 3단계 웨이블릿 변환시 기존의 방법의 약 15.6%의 계산량으로 움직임 추정을 한다. 그리고 부호화시 움직임 벡터가 각 부대역마다 존재하게 되므로 움직임 벡터가 약 7배 늘어나게 되지만, 더 세밀한 움직임 추정을 할 수 있게 되므로 움직임 보상 예측 오차의 부호화량이 줄어들게 되어 부호화 효율이 기존의 방법보다 좋아지게 된다. 압축을 하지 않았을 경우 평균 MAD면에서 약 0.3∼11.6% 가량 개선되었고, 압축을 할 때 동일한 비트율에서 PSNR이 약 0.3∼3.0㏈ 정도 개선되었다.

Keywords

References

  1. Coded Representation of Picture and Audio Information Test Model 5 ISO/IEC JTC1/SC29/WG11, MPEG93/457
  2. MPEG-4 Video Verification Model Version 10.0 ISO/IEC JTCI/SC29/WG11 N1992
  3. Draft-T ITU-T Recommendation H.263
  4. IEEE Trans. Information Theory 36 The Wavelet Transform, Time-Frequency Localization and Signal Analysis I.Daubechies
  5. Commun. on Pure and Appl. Math. v.41 Orthonormal Bases of Compactly Supported Wavelets I.Daubechies
  6. IEEE Trans. Pattern Anal. Machine Intel. v.11 A Theory for Multiresolution Signal Decomposition: The Wavelet Repres-entation S.Mallat
  7. IEEE Trans. Image Proc. v.1 Image Coding Using the Wavelet Transform M.Antonini;M.Barlaud;P.Mathieu;I.Daubechies
  8. IEEE Trans. Signal Proc. v.41 Embedded Image Coding Using Zerotree Wavelets Coeffients J.Shapiro
  9. IEEE Trans. Circuits and Systems for Video Technology v.8 Inter-frame Coding Using Two-Stage Variable Block-Size Multiresolution Motion Estimation and Wavelet Decomposition Kim S. Rhee,J.Jeon;K,Park
  10. Proc. SPIE Visual Communications Image Processing v.3309 Adaptive Mutiresolution Motion Estimation Techniques for Wavelet-Based Video Coding S.Kim;T.Aboulnasr;S.Panchanathan
  11. Optical Engineering v.35 Multiresolution Motion Estimation Techniques for Video Compression M.Mandal;E.Chan;X.Wang;S.Pan chanathan
  12. IEEE Trans. on Circuits And Systems for Video Technology v.2 no.3 Motion-Compensated Wavelet Transform Coding for Color Video Compression Y.Zhang;S.Zafar
  13. IEEE Trans. On CSVT v.1 no.1 Interpolative Multiresolution Coding of Advanced Television with Compatible Subchannels K.Uz;M.Vetterli;D.Legall
  14. Proc. of ICASSP Multiresolution Motion Estimation G.Conklin;S.Hemami
  15. IEEE Trans. on Image Processing v.9 no.4 Motion Estimation Using Low-Band-Shift Method for Wavelet-Based Moving-Picture Coding H.Park;H.Kim
  16. Image and Vide Compression Standards V.Bhaskaran;K.Konstantinides
  17. Image and Video Compression for Multimedia Engineering Y.Shi;H.Sun
  18. JPEG2000 Image Coding System, Ver 1.0 ISO/IEC CD15444-1
  19. JBIG2 Final Committee Draft ISO/IEC JTCI/SC29/WGI (ITU-TSG 8), N1545,