摘要
H.264/AVC是新一代视频压缩标准。在其编码技术中,运动估计是提高编码效率的关键。在简单描述块匹配运动估计原理的基础上,就运动模板选择对提高运动估计算法的效率方面作了重点论述,提出了一种新的基于运动特征的自适应运动估计算法。该算法首先建立具有自适应特性的搜索起点预测模型,采用的搜索模板可以根据物体的运动特征调整形状,从而提高搜索效率。实验表明,新算法在图像质量和搜索速度方面有较好的优势。
The H. 264/ACV is a kind of new generation video compression standard. In its coding technique, motion estimation is the key in increasing coding efficiency. The block-matching motion estimation principle is briefly described in this paper,based on that,we discuss in detail the function of motion template selection on improving the efficiency of motion estimation algorithms, and put forward a new adaptive mo- tion estimation algorithm based on motion characteristics. The algorithm sets up an adaptive model for starting search point prediction first, and the search template used can adjust its shape according to the motion characteristics, so as to improve searching efficiency. Simulation experi- ments show that the new algorithm is superior in image quality and search speed.
出处
《计算机应用与软件》
CSCD
2010年第8期145-147,共3页
Computer Applications and Software
基金
辽宁省自然科学基金项目(20061025)
