H.264/AVC Fast Mode Decision Using Pattern Recognition and Swarm Intelligence

• Main Authors: Heng-Yi Jhu, 朱恒逸
• Other Authors: Jia-ching Wang
• Format: Others
• Language: zh-TW
• Published: 2012
• Online Access: http://ndltd.ncl.edu.tw/handle/83584551453075045334

碩士 === 國立中央大學 === 資訊工程研究所 === 100 === Video coding has been widely applied to TV transmission, network stream, monitoring system, video conferencing, and multimedia devices. In the video coding standard H.264/AVC, several coding techniques are added. Compared to old Standards such as MPEG-2 and H.263, H.264/AVC provides better coding efficiency. It raises the compression ability and enhances the quality of pictures. But it pays a huge computation time for it, making H.264/AVC cannot be realized easily in the real time coding environment. So, reducing the huge coding complexity with good compression quality has been a hot issue. This thesis focuses on speeding up both the inter coding and intra coding procedures. We propose two efficient mode decision algorithms for intra coding and inter coding. They greatly reduce the coding time but keep superior coding performance. After our experiments, the approaches we proposed speed up intra coding and inter coding by 60% and 46%, respectively. Besides, its coding performance is very close to original H.264/AVC.