Translate this page:
Please select your language to translate the article


You can just close the window to don't translate
Library
Your profile

Back to contents

Software systems and computational methods
Reference:

Arzumanyan R.V., Sukhinov A.I. Factorization of the 8x8 HEVC Video Coding Inverse Discrete Transform Matrix and the Fast Algorithm Thereupon

Abstract: The subject of the present research is the development of the algorithm for the fast inverse discrete transform of the 8x8 ITU-T H265 (HEVC) video coding standard. The authors of the article analyze differences between the structures of the inverse transform matrix and inverse discrete cosine transform matrix as well as approaches that may be applied to factorization of the aforesaid matrix. They also provide an evaluation of a number of operations necessary to perform the transfer. The authors conduct a numerical experiment to prove the efficiency of the developed algorithm from the point of view of the speed of performance on the central processing unit (CPU). The research method used by the authors is the theoretical analysis and numerical experiment including collection of relevant information and analysis of results. To conduct the numerical experiment the authors have written a C-language program that executes a standard algorithm of the inverse transfer (direct multiplication of the transform matrix and coefficient vectors) and the fast algorithm of the inverse transfer as it is described in the theoretical part of the research. Then the authors have compared the productivity results. The novelty of the research is caused by the fact that the authors offer a new algorithm for the fast transfer of the 8x8 HEVC standard and the scheme of inverse matrix factorization. Compared to previous researches and algorithm, the given algorithm requires fewer arithmetic operations, thus takes less time. At the end of their research article the authors make conclusions regarding the possibility of the fast inverse transfer of the HEVC standard, offer their own scheme for the aforesaid 8x8 matrix factorization and develop the fast algorithm for the inverse transfer based on the discovered factorization schemes. 


Keywords:

algorithm analysis, lossy compression, codec, video compression, inverse discrete transform, matrix factorization, fast algorithm, HEVC, software codec, hardware codec


This article can be downloaded freely in PDF format for reading. Download article


References
1. Ma T., Liu C., Yibo F., Zeng X. A fast 8x8 IDCT algorithm for HEVC // V kn.: IEEE ASIC (ASICON). 2013. S. 203-208.
2. Cooley J.W., Tukey J.W. An Algorithm for the Machine Calculation of Complex Fourier Series // Mathematics of Computation. 1965. № 90. S. 297-301.
3. Bleykhut R. Bystrye algoritmy tsifrovoy obrabotki signalov. M.: Mir, 1989. 448 s.
4. Chen W.-H., Harrison-Smith C., Fralick S. C. A Fast Computational Algorithm for the Discrete Cosine Transform // IEEE transactions on communications. 1977. № 9. S. 1004-1009.
5. Winken M., Helle P., Marpe D., Schwarz H., Wiegand T. Transform coding in the HEVC test model // V kn.: IEEE International Conference on Image Processing. 2011. S. 3693–3696.
6. Park J.-S., Nam W.-J., Han S.-M., Lee S. 2-D Large Inverse Transform (16x16, 32x32) for HEVC (High Efficiency Video Coding) // Journal of semiconductor technology and science. 2012. № 12. S. 204-208.
7. Hung C.-Y., Landman P. Compact inverse discrete cosine transform circuit for MPEG video decoding // V kn.: IEEE SIPS. 1997. S. 364–373.
8. Budagavi M., Fuldseth A., Bjøntegaard G., Sze V., Sadafale M. Core Transform Design in the High Efficiency Video Coding (HEVC) Standard // IEEE Journal of Selected Topics in Signal Processing. 2013. № 6. S. 1029-1041.
9. Belghith F., Loukil H., Masmoudi N. Efficient Hardware Architecture of a Modified 2‐D Transform for the HEVC Standard // International Journal of Computer Science and Application. 2013. № 4. S. 59-69.
10. Budagavi M,, Sze V. Unified forward+inverse transform architecture for HEVC // V kn.: IEEE International Conference on Image Processing. 2012. S. 209-212.