CABAC在H.264/AVC中的应用  

Application of Context-bsaed Adaptive Binary Arithmetic Coding in H.264/AVC

作　　者：闵玲[1] 李方[1] 何小海[1] 

机构地区：[1]四川大学电子信息学院,四川成都610065

出　　处：《信息与电子工程》2006年第4期269-274,共6页information and electronic engineering

摘　　要：H.264/AVC是由国际电信联盟和国际标准化组织共同制定的新一代视频编码标准。在该标准中,规定了两种熵编码的模式,即基于上下文的自适应二进制算术编码(Context-bsaedAdaptiveBinaryArithmeticCoding,CABAC)和基于上下文自适应可变长编码(Context-bsaedAdaptiveVariable-LengthCoding,CAVLC)。其中,CABAC作为一种新型的熵编码方法,将自适应技术、上下文模型化和二进制算术编码有地的结合在一起,达到了较高的压缩效率,CABAC的框架中还使用了一些新颖的方法,使得CABAC在软硬件的实现上更加方便。为了验证CABAC的实际效果,笔者应用参考程序对其进行了直观的测试,实验结果表明:在相同图像质量下,CABAC和CAVLC相比的确能节省较大的平均比特率。The emerging H.264 / AVC video coding standard is developed collaboratively by the ITU-T and ISO/IEC. The standard specifies two types of entropy coding：Context-based Adaptive Binary Arithmetic Coding （CABAC） and Context-bsaed Adaptive Variable-Length Coding （CAVLC）. By effectively combining adaptation technique, context model and binary arithmetic coding together, CABAC as a new entropy coding method has achived a high level compression. There are some new ways to make it easy to be carried out. To prove the performance of CABAC,we test it by referenced application. The experiment indicates that CABAC saves bit rate than CAVLC with the same quality of image.

关 键 词：H.264/AVC 算术编码 CABAC 二值化 上下文模型 概率模型 

分 类 号：TN919.81[电子电信—通信与信息系统]