An analysis of the DCT coefficient distribution with the H.264 video coder

By extensive analysis with several video sequences, we observed that the statistical distribution of the DCT coefficients in typical video coding applications is closer to a Cauchy distribution than to a Laplacian distribution. We developed the rate and the distortion expressions as a function of the video coder quantization parameter based on this observation. Experiments with an H.264 codec demonstrate that the Cauchy distribution based expressions provide better estimates for the actual rate and distortion than those that are based on the Laplacian distribution.

[1]  Ajay Luthra,et al.  Overview of the H.264/AVC video coding standard , 2003, IEEE Trans. Circuits Syst. Video Technol..

[2]  Gregory K. Wallace,et al.  The JPEG still picture compression standard , 1992 .

[3]  Joseph W. Goodman,et al.  A mathematical analysis of the DCT coefficient distributions for images , 2000, IEEE Trans. Image Process..

[4]  Jerry D. Gibson,et al.  Distributions of the Two-Dimensional DCT Coefficients for Images , 1983, IEEE Trans. Commun..

[5]  Lawrence A. Rowe,et al.  DCT coefficient distributions , 1996, Electronic Imaging.

[6]  Hocine Cherifi,et al.  On the distribution of the DCT coefficients , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[7]  A.N. Netravali,et al.  Picture coding: A review , 1980, Proceedings of the IEEE.

[8]  Gary J. Sullivan,et al.  Rate-constrained coder control and comparison of video coding standards , 2003, IEEE Trans. Circuits Syst. Video Technol..