一种基于三维小波系数上下文模型的视频压缩方法

受视频编码方法3D-ESCOT的启发，提出了一种新的基于三维小波系数上下文模型的视频编码方法．该方法分为：首先．利用人眼视觉模型（HVS）对经过三维小波变换的小波系数进行量化；然后利用空间和时间相邻的系数对当前系数进行预测．以预测值的量化值作为上下文，对系数进行基于上下文的算术编码．该方法具有空间和时间分辨率可扩展性质，在视频传输中能够自动适应不同类型设备和网络带宽的变化．实验表明，该方法不仅在每个质量等级和每个时间和空同分辨率获得了比3D-ESCOT更高的压缩比，而且编解码所需要的时间更少．

Video Coding Scheme Based on the Context Model of Three-Dimensional Wavelet Coefficients

Inspired by the 3D-ESCOT video coding scheme,a video coding scheme based on the context model of three-dimensional wavelet coefficients is presented. The scheme includes two stages: At the first stage,wavelet coefficients resulting from three-dimensional wavelet transform are quantized by the threshold formed by the eye's sensitivity to their subbands.At the second stage,the wavelet coefficients are encoded by the entropy encoder,with the contexts formed by quantizing their prediction values,which are determined by their neighbors in space and time dimensions.The scheme is resolution scalable in time and space,which can adapt to heterogeneous devices and networks.Furthermore,experimental results show that for every quantization width,every scale in time and every scale in space,the scheme not only achieves higher compression rate but also costs less time than 3D-ESCOT.