Share Email Print

Proceedings Paper

A fast pattern chosen algorithm for intra-frames prediction
Author(s): Qiwei Lin; Gui Feng
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

It is well known that in the H.264 video coding standard the image inner spatial correlation had been made full use in its intra-frames prediction algorithm, so the encoding efficiency can be improved. However, in H.264 video coding standard, no matter prediction frames (P-frame or B-frame) or intra-frames (I-frame), all of them need to carry on predicting at each macro block inside the frame, and the algorithm utilizes the surrounding pixels to predict, therefore it improves the encoding efficiency at the cost of increasing calculation complexity of the encoder. For the sake of real time video communications, it is necessary to employ a new algorithm to reduce calculating complexity for intra-frames predicting and speed up encoding procedure. For these reasons, a modified fast intra-frames prediction pattern decision algorithm was proposed in this paper. Using directional information of luminance pattern and chrominance pattern to decide the candidate pattern, the limited of edge direction can restrict the number of the candidate pattern, so the algorithm can greatly decrease the number of candidate patterns, therefore it can obviously cut down calculating amount. At the same time, the proposed algorithm also adopts early termination of 4×4 block pattern selection strategy to choose the most possible pattern. The proposed algorithm has the ability of obviously reducing the pattern selecting calculation complexity, which has been approved by the experimental result. So the algorithm is possible to be used in the real time video communication.

Paper Details

Date Published: 19 September 2007
PDF: 9 pages
Proc. SPIE 6683, Satellite Data Compression, Communications, and Archiving III, 66830O (19 September 2007); doi: 10.1117/12.740114
Show Author Affiliations
Qiwei Lin, Hua Qiao Univ. (China)
Gui Feng, Hua Qiao Univ. (China)

Published in SPIE Proceedings Vol. 6683:
Satellite Data Compression, Communications, and Archiving III
Roger W. Heymann; Bormin Huang; Irina Gladkova, Editor(s)

© SPIE. Terms of Use
Back to Top