A New Framework for Distributed Video Coding Based on JPEG 2000
A New Framework for Distributed Video Coding Based on JPEG 2000
复制标题
DOI：
10.1109/icassp.2006.1660639
复制
发表时间：
2006-05
期刊：
影响因子：
--
通讯作者：
Y. Tonomura;T. Nakachi
中科院分区：
文献类型：
--
作者：
Y. Tonomura;T. Nakachi
分布式视频编码（DVC）是基于Slepian-Wolf和Wyner-Ziv提出的定理，作为一种新的视频压缩范式而备受关注。在DVC系统中，几个编码器将发送比特流到一个解码器，该解码器必须处理所有传入的比特流。一些DVC系统使用基于DCT的帧内压缩。然而，传统的DVC系统与DCT的亲和力较低，因为它们在解压缩所有比特流之前无法生成必要的侧信息。在本文中，我们提出了一种新的DVC方案，该方案是在解压缩所有比特流之前生成侧信息的一种简单方法。该方案利用了jpeg2000的可扩展性和jpeg2000第二部分的多分量变换。测试结果表明，新方案的PSNR比传统的JPEG 2000高约7[dB]
Distributed video coding (DVC), based on the theorems proposed by Slepian-Wolf and Wyner-Ziv, is attracting attention as a new paradigm for video compression. In DVC systems, several encoders will send bit streams to a single decoder which must handle all incoming bit streams. Some of the DVC systems use intraframe compression based on DCT. However, conventional DVC systems have low affinity with DCT, because they fail to generate the necessary side information until after decompressing all bit streams. In this paper, we propose a new DVC scheme that is an easy way to generate the side information before decompressing all bit streams. The scheme utilizes scalability of JPEG 2000 and the multicomponent transforms of JPEG 2000 part-2. Tests confirm that the PSNR of the new scheme is about 7[dB] higher than that of conventional JPEG 2000