摘要： 传统的转码技术在H.264到高效视频编码(HEVC)视频的转码过程中效率低下。针对该问题,提出一种利用H.264解码信息加速HEVC编码的转码算法。该算法利用纹理复杂度和编码单元所含比特数之间的关系,根据每个编码单元所含比特数和每帧平均比特数的比值自适应地预 测编码搜索深度范围,对H.264中宏块的运动矢量进行预处理,减少预测单元分割模式的候选数目,并优化运动搜索起点和运动搜索范围。实验结果表明,与HEVC相比,该算法能缩短编码时间。

关键词: H.264标准, 转码, 运动矢量, 模式选择, 运动估计

Abstract: Traditional transcoding technology form H.264 to High Efficiency Video Coding (HEVC) has a very low coding efficiency.In order to solve this problem,an algorithm using decoding information from H.264 to accelerate encoding of HEVC is proposed.This paper adaptively predicts the range of encoding searching depth with the number of bits in each coding unit based on the relationship between texture complexity and coding bits,pre-processes the motion vector abstracted from H.264 to reduce the candidates of prediction mode and optimize the motion estimation with motion vector predictor and search window.Experimental results show that the proposed algorithm can reduce encoding time compared with HEVC.

Key words: H.264 standard, transcoding, motion vector, mode selection, motion estimation

中图分类号: 

• TP391