Structured Vector Quantization Theory

结构化矢量量化理论

• 批准号: 9415754
• 负责人: David Neuhoff
• 金额: $ 37.37万
• 依托单位: Regents of the University of Michigan - Ann Arbor
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 1995
• 资助国家: 美国
• 起止时间: 1995-06-01 至 1999-11-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9415754&HistoricalAwards=false
• 关键词: Structured; Vector; Quantization; Theory

9415754 Neuhoff This project addresses the development of the theory and practice ofstructured vector quantizers for signal compression of sources such as speech, images, video and music. Though the cost of storing and transmitting information continues to decrease, such reductions seem only to stimulate demand for information and, consequently, for further developments in compression. Structured vector quantizers from a class of compression techniques that hold significant promise for gains in performance. This project will develop theory to increase the qualitative and quantitative understanding of the performance of structured vector quantizers, particularly the tradeoffs between performance and complexity. It will analyze a number of specific structured vector quantizers. In addition, it will develop new structured quantization approaches. The methodology is largely, but not exclusively, based on asymptotic quantization theory techniques and concepts. The analysis and developments are for fixed and variable-rate quantizers that minimize rate with respect to squared error and perceptual fidelity criteria. ***

小行星9415754 本计画致力于发展结构向量量化器的理论与实务，以压缩语音、影像、视讯与音乐等来源的讯号。 虽然存储和传输信息的成本不断降低，但这种降低似乎只会刺激对信息的需求，从而促进压缩技术的进一步发展。 来自一类压缩技术的结构化矢量量化器，具有显著的性能增益承诺。 这个项目将发展理论，以增加对结构化矢量量化器性能的定性和定量理解，特别是性能和复杂性之间的权衡。 它将分析一些具体的结构化矢量量化器。 此外，它将开发新的结构化量化方法。该方法在很大程度上，但不完全是基于渐近量子化理论的技术和概念。的分析和发展是固定和可变速率的量化器，最大限度地减少速率平方误差和感知保真度标准。 ***