融合宏块平坦度和时空相关性的预测模式选择算法

Prediction mode decision algorithm combining flatness and spatial-temporal correlation of macro-block

H.264/AVC以巨大编码复杂度为代价,在获得更高压缩率的同时,编码实时性也随之降低。针对视频编码中重要且耗时的帧间预测技术,分析了宏块平坦度和时空相关性,提出了一种快速的预测模式选择算法。仿真实验结果表明,本文提出算法与H.264/AVC(JM12.2)标准算法相比,在保持重建视频图像质量和输出码流结构的前提下,平均节省约70%的编码时间,并继承了H.264/AVC低码率的编码优势。

At the cost of increased computational complexity, as H.264/AVC obtains higher compression ratio, its encoding real-time performance also fall subsequently. Aiming at the important and time-consuming inter-frame prediction technique in video encoding, flatness characteristics and spatio-temporal correlation of macro-block are analyzed. A fast inter-frame prediction mode decision algorithm is proposed Simulation results show that the proposed algorithm can reduce about 70% on average of the total encoding time compared with H.264/AVC （JM12.2）, with keeping video image quality and reconstruction output streaming structure, inheriting low bit-rate coding advantage.