• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >End-to-end rate distortion analysis and optimization for robust video transmission over lossy networks.
【24h】

End-to-end rate distortion analysis and optimization for robust video transmission over lossy networks.

机译：端到端速率失真分析和优化，用于有损网络上的稳健视频传输。

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This dissertation is concerned with enhancing the error resilience of video transmission over lossy networks, such as the Internet and wireless networks. The major objective is to balance the tradeoff between efficiency and robustness within a rate-distortion (RD) framework, so that the best reconstruction quality is achieved under the prescribed bit rate budget and network constraints.; The overall quality of video transmission over lossy networks is an intricate function of several factors including the source and channel coding schemes, the network status, and the decoder's recovery actions. This fact motivates the derivation of the proposed “end-to-end” approach.; The first part of the dissertation focuses on the system performance analysis and the derivation of efficient algorithms to estimate the end-to-end distortion. The live video coding and pre-compressed video streaming scenarios are addressed separately, due to the fundamental difference in the level of network awareness during compression, which hence poses different challenges and requires different solutions. A recursive optimal per-pixel estimate (ROPE) is proposed for calculating, at the encoder, the expected end-to-end distortion of live video coding and transmission over lossy networks. For the streaming of fixed pre-compressed video, a first order distortion estimation (FODE) algorithm is derived, which is applicable to virtually all source encoding techniques, network settings, varying channel conditions, and specific error-resilience schemes. Both algorithms take into account the effects of quantization, packet loss, spatial and temporal error propagation, and error concealment. They feature high accuracy and low complexity, and are compatible with video coding standards.; Given the above capacity for accurate end-to-end distortion analysis, the second part of the dissertation is concerned with balancing the tradeoff between efficiency and robustness within a rate-distortion framework. The control parameters in a given system are optimally selected to explicitly and directly minimize the end-to-end distortion. We first propose a technique of RD-optimized adaptive intra-update at the MB level to optimize the robustness of the live video transmission. The technique is then extended to handle the standard layered coding system, including enhancement of both encoder-based and decoder-based error-resilience measures. We finally propose an RD-optimized adaptation solution for the robust streaming of pre-compressed video. Substantial gains in the simulation results demonstrate the superiority of the proposed end-to-end RD-optimized approaches.
机译：本发明涉及增强诸如因特网和无线网络之类的有损网络上的视频传输的错误恢复能力。主要目标是在速率失真（RD）框架内平衡效率和鲁棒性之间的权衡，以便在规定的比特率预算和网络约束条件下获得最佳的重构质量。有损网络上视频传输的整体质量是多种因素的复杂功能，其中包括源和信道编码方案，网络状态以及解码器的恢复动作。这一事实促使了所提议的“端到端”方法的推导。论文的第一部分着重于系统性能分析以及推导端到端失真的高效算法的推导。实时视频编码和预压缩视频流方案是分别解决的，这是由于压缩过程中网络意识水平的根本差异，因此带来了不同的挑战，并需要不同的解决方案。提出了一种递归最佳每像素估计（ROPE），用于在编码器处计算实时视频编码和在有损网络上传输的预期端到端失真。对于流式传输固定的预压缩视频，需要导出一阶失真估计（FODE）算法，该算法实际上适用于所有源编码技术，网络设置，变化的信道条件和特定的容错方案。两种算法都考虑了量化，包丢失，空间和时间错误传播以及错误隐藏的影响。它们具有高精度和低复杂度的特点，并且与视频编码标准兼容。考虑到上述精确的端到端失真分析能力，本文的第二部分涉及在速率失真框架内平衡效率和鲁棒性之间的权衡。最佳选择给定系统中的控制参数，以明确，直接地减小端到端失真。我们首先提出一种MB级别的RD优化自适应帧内更新技术，以优化实时视频传输的鲁棒性。然后将该技术扩展为处理标准的分层编码系统，包括增强基于编码器和基于解码器的容错措施。最后，我们为预压缩视频的鲁棒流提供了RD优化的自适应解决方案。仿真结果的显着提高证明了所提出的端到端RD优化方法的优越性。

著录项

  • 作者

    Zhang, Rui.;

  • 作者单位

    University of California, Santa Barbara.;

  • 授予单位 University of California, Santa Barbara.;
  • 学科 Engineering Electronics and Electrical.
  • 学位 Ph.D.
  • 年度 2001
  • 页码 140 p.
  • 总页数 140
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 无线电电子学、电信技术;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号