摘要
为了降低高效视频编码(HEVC)的帧内编码计算复杂度,本文提出一种应用于预测模式决策和编码单元大小决策的帧内模式快速编码组合算法.在预测模式选择上,本文根据预测单元(PU)的纹理信息对粗模式决策(RMD)和率失真最优化(RDO)候选列表进行筛选以减少计算复杂度.在编码单元(CU)大小决策方面,根据CU的纹理复杂度,对平滑简单的CU自适应地终止划分,对纹理复杂的CU跳过编码提前进行划分,减少CU深度遍历的时间.实验结果表明,本文算法与HM参考模型相比,能够平均降低38. 0%的编码时间,而BDBR仅增加0. 33%.与现有的优秀算法相比,本文算法进一步降低了编码时间,且具有更好的编码质量.
To reduce the computational complexity of high efficiency video coding( HEVC) Intra Coding,a fast intra mode coding algorithm is proposed by exploiting prediction mode decision and coding unit( CU) size decision in this paper. For prediction mode decision,the algorithm eliminates the unlikely prediction modes and candidates for Rough Mode Decision( RMD) and rate-distortion optimization( RDO) to reduce the computational complexity,respectively,according to the textural feature in the current prediction unit( PU). For CU size decision,a texture-based method is proposed. For the homogenous regions,CUs terminate the division adaptively.For the complicate regions,CUs skip the coding process and split early to reduce the depth traversal time. Compared to the HEVC test Model,experimental results showthat the proposed method reduces 38. 0% encoding time on average with only 0. 33% BD-rate increase. And compared with the state-of-art methods,the proposed algorithm makes progress in time reduction and has a better performance of coding quality.
作者
易清明
谢志煌
石敏
YI Qing-ming;XIE Zhi-huang;SIN Min(College of Information Science and Technology,Jinan University,Guangzhou 510632,China)
出处
《小型微型计算机系统》
CSCD
北大核心
2019年第1期199-204,共6页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61603153)资助
广州市科技计划项目(201604016085)资助