Fast Intermode Decision Via Statistical Learning for H.264 Video Coding

  • Wei-Hau Pan
  • Chen-Kuo Chiang
  • Shang-Hong Lai
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4903)

Abstract

Although the variable-block-size motion compensation scheme significantly reduces the compensation error, the computational complexity of motion estimation (ME) is tremendously increased at the same time. To reduce the complexity of the variable-block-size ME algorithm, we propose a statistical learning approach to simplify the computation involved in the sub-MB mode selection. Some representative features are extracted during ME with fixed sizes. Then, an off-line pre-classification approach is used to predict the most probable sub-MB modes according to the run-time features. It turns out that only possible sub-MB modes need to perform ME. Experimental results show that the computation complexity is significantly reduced while the video quality degradation and bitrate increment is negligible.

Keywords

Variable block-size H.264 motion estimation statistical learning 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Wu, D., Pan, F., Lim, K.P., Wu, S., Li, Z.G., Lin, X., Rahardja, S., Ko, C.C.: Fast Intermode Decision in H.264/AVC Video Coding. IEEE Trans. Circuits & Systems For Video Technology 15(6), 953–958 (2005)CrossRefGoogle Scholar
  2. 2.
    Kuo, T.-Y., Chan, C.-H.: Fast Variable Block Size Motion Estimation for H.264 Using Likelihood and Correlation of Motion Field. IEEE Trans. Circuits & Systems for Video Technology 16(10), 1185–1195 (2006)CrossRefGoogle Scholar
  3. 3.
    Jain, A.K.: Fundamentals of digital image processing, ch. 4. Prentice-Hall, Englewood Cliffs (1989)MATHGoogle Scholar
  4. 4.
    Tourapis, A.M.: Enhanced Predictive Zonal Search for Single and Multiple Frame Motion Estimation. In: SPIE Proceedings of the Visual Comm. Image Proc. (2002)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Wei-Hau Pan
    • 1
  • Chen-Kuo Chiang
    • 1
  • Shang-Hong Lai
    • 1
  1. 1.Department of Computer Science, National Tsing Hua University, HsinchuTaiwan

Personalised recommendations