摘要
深入分析了H.264标准中CAVLC编解码方法,在此基础上针对参考实现中的遍历查找算法搜索效率不高的问题,引入了子表法和二叉树-子表混合法两种算法来提高解码效率。子表法是将码表分成若干张子表,再在子表中进行搜索;二叉树-子表法结合子表划分,在子表内部采用基于二叉树的形式解码。实验结果表明,在内存增加不多的情况下,这两种算法比参考实现原有算法的搜索速度分别提高了1.7倍和3倍。
Based on thorough analysis of coding and decoding methods of CAVLC in H.264, this paper proposes two methods, subtable method and hybrid method from binary tree and sub table, to improve the normal table look-up decoding efficiency implemented in reference code. Sub-table method splits the original table into several small ones for improving search efficiency, and binary tree & sub table hybrid method employs binary tree within each sub table. The testing result shows that these two methods can provide 1.7 and 3 times higher speed individually compared with the original one at the cost of only a little memory space.
出处
《微计算机信息》
北大核心
2006年第09Z期25-27,共3页
Control & Automation
基金
上海市自然科学基金:(04ZR14082)