A fast encoding algorithm based on the mean square error （MSE） distortion for vector quantization is introduced. The vector, which is effectively constructed with wavelet transform （WT） coefficients of images, can...A fast encoding algorithm based on the mean square error （MSE） distortion for vector quantization is introduced. The vector, which is effectively constructed with wavelet transform （WT） coefficients of images, can simplify the realization of the non-linear interpolated vector quantization （NLIVQ） technique and make the partial distance search （PDS） algorithm more efficient. Utilizing the relationship of vector L2-norm and its Euclidean distance, some conditions of eliminating unnecessary codewords are obtained. Further, using inequality constructed by the subvector L2-norm, more unnecessary codewords are eliminated. During the search process for code, mostly unlikely codewords can be rejected by the proposed algorithm combined with the non-linear interpolated vector quantization technique and the partial distance search technique. The experimental results show that the reduction of computation is outstanding in the encoding time and complexity against the full search method.展开更多
In this paper a novel coding method based on fuzzy vector quantization for noised image with Gaussian white-noise pollution is presented. By restraining the high frequency subbands of wavelet image the noise is signif...In this paper a novel coding method based on fuzzy vector quantization for noised image with Gaussian white-noise pollution is presented. By restraining the high frequency subbands of wavelet image the noise is significantly removed and coded with fuzzy vector quantization. The experimental result shows that the method can not only achieve high compression ratio but also remove noise dramatically.展开更多
In this paper, a new amplitude quantization synthesis method for ultralow sidelobe phased arrays is proposed, which is based on the constrained nonlinear optimization algorithm. By introducing a set of critical constr...In this paper, a new amplitude quantization synthesis method for ultralow sidelobe phased arrays is proposed, which is based on the constrained nonlinear optimization algorithm. By introducing a set of critical constraint conditions into the optimization model, we can directly quantize the amplitude distribution instead of replacing it with a continuous equivalent aperture antenna. The mutual coupling and the element patterns are also considered in the quantization synthesis. Finally, some array simulation results are given to show the effectiveness of the method.展开更多
随着人工智能的发展,深度神经网络成为多种模式识别任务中必不可少的工具,由于深度卷积神经网络(CNN)参数量巨大、计算复杂度高,将它部署到计算资源和存储空间受限的边缘计算设备上成为一项挑战。因此,深度网络压缩成为近年来的研究热...随着人工智能的发展,深度神经网络成为多种模式识别任务中必不可少的工具,由于深度卷积神经网络(CNN)参数量巨大、计算复杂度高,将它部署到计算资源和存储空间受限的边缘计算设备上成为一项挑战。因此,深度网络压缩成为近年来的研究热点。低秩分解与向量量化是深度网络压缩中重要的两个研究分支,其核心思想都是通过找到原网络结构的一种紧凑型表达,从而降低网络参数的冗余程度。通过建立联合压缩框架,提出一种基于低秩分解和向量量化的深度网络压缩方法——可量化的张量分解(QTD)。该方法能够在网络低秩结构的基础上实现进一步的量化,从而得到更大的压缩比。在CIFAR-10数据集上对经典ResNet和该方法进行验证的实验结果表明,QTD能够在准确率仅损失1.71个百分点的情况下,将网络参数量压缩至原来的1%。而在大型数据集ImageNet上把所提方法与基于量化的方法PQF(Permute,Quantize,and Fine-tune)、基于低秩分解的方法TDNR(Tucker Decomposition with Nonlinear Response)和基于剪枝的方法CLIP-Q(Compression Learning by In-parallel Pruning-Quantization)进行比较与分析的实验结果表明,QTD能够在相同压缩范围下实现更好的分类准确率。展开更多
预训练模型通过自监督学习表示在非平行语料语音转换(VC)取得了重大突破。随着自监督预训练表示(SSPR)的广泛使用,预训练模型提取的特征中被证实包含更多的内容信息。提出一种基于SSPR同时结合矢量量化(VQ)和联结时序分类(CTC)的VC模型...预训练模型通过自监督学习表示在非平行语料语音转换(VC)取得了重大突破。随着自监督预训练表示(SSPR)的广泛使用,预训练模型提取的特征中被证实包含更多的内容信息。提出一种基于SSPR同时结合矢量量化(VQ)和联结时序分类(CTC)的VC模型。将预训练模型提取的SSPR作为端到端模型的输入,用于提高单次语音转换质量。如何有效地解耦内容表示和说话人表示成为语音转换中的关键问题。使用SSPR作为初步的内容信息,采用VQ从语音中解耦内容和说话人表示。然而,仅使用VQ只能将内容信息离散化,很难将纯粹的内容表示从语音中分离出来,为了进一步消除内容信息中说话人的不变信息,提出CTC损失指导内容编码器。CTC不仅作为辅助网络加快模型收敛,同时其额外的文本监督可以与VQ联合优化,实现性能互补,学习纯内容表示。说话人表示采用风格嵌入学习,2种表示作为系统的输入进行语音转换。在开源的CMU数据集和VCTK语料库对所提的方法进行评估,实验结果表明,该方法在客观上的梅尔倒谱失真(MCD)达到8.896 d B,在主观上的语音自然度平均意见分数(MOS)和说话人相似度MOS分别为3.29和3.22,均优于基线模型,此方法在语音转换的质量和说话人相似度上能够获得最佳性能。展开更多
基金the National Natural Science Foundation of China (60602057)the NaturalScience Foundation of Chongqing Science and Technology Commission (2006BB2373).
文摘A fast encoding algorithm based on the mean square error （MSE） distortion for vector quantization is introduced. The vector, which is effectively constructed with wavelet transform （WT） coefficients of images, can simplify the realization of the non-linear interpolated vector quantization （NLIVQ） technique and make the partial distance search （PDS） algorithm more efficient. Utilizing the relationship of vector L2-norm and its Euclidean distance, some conditions of eliminating unnecessary codewords are obtained. Further, using inequality constructed by the subvector L2-norm, more unnecessary codewords are eliminated. During the search process for code, mostly unlikely codewords can be rejected by the proposed algorithm combined with the non-linear interpolated vector quantization technique and the partial distance search technique. The experimental results show that the reduction of computation is outstanding in the encoding time and complexity against the full search method.
文摘In this paper a novel coding method based on fuzzy vector quantization for noised image with Gaussian white-noise pollution is presented. By restraining the high frequency subbands of wavelet image the noise is significantly removed and coded with fuzzy vector quantization. The experimental result shows that the method can not only achieve high compression ratio but also remove noise dramatically.
文摘In this paper, a new amplitude quantization synthesis method for ultralow sidelobe phased arrays is proposed, which is based on the constrained nonlinear optimization algorithm. By introducing a set of critical constraint conditions into the optimization model, we can directly quantize the amplitude distribution instead of replacing it with a continuous equivalent aperture antenna. The mutual coupling and the element patterns are also considered in the quantization synthesis. Finally, some array simulation results are given to show the effectiveness of the method.
文摘随着人工智能的发展,深度神经网络成为多种模式识别任务中必不可少的工具,由于深度卷积神经网络(CNN)参数量巨大、计算复杂度高,将它部署到计算资源和存储空间受限的边缘计算设备上成为一项挑战。因此,深度网络压缩成为近年来的研究热点。低秩分解与向量量化是深度网络压缩中重要的两个研究分支,其核心思想都是通过找到原网络结构的一种紧凑型表达,从而降低网络参数的冗余程度。通过建立联合压缩框架,提出一种基于低秩分解和向量量化的深度网络压缩方法——可量化的张量分解(QTD)。该方法能够在网络低秩结构的基础上实现进一步的量化,从而得到更大的压缩比。在CIFAR-10数据集上对经典ResNet和该方法进行验证的实验结果表明,QTD能够在准确率仅损失1.71个百分点的情况下,将网络参数量压缩至原来的1%。而在大型数据集ImageNet上把所提方法与基于量化的方法PQF(Permute,Quantize,and Fine-tune)、基于低秩分解的方法TDNR(Tucker Decomposition with Nonlinear Response)和基于剪枝的方法CLIP-Q(Compression Learning by In-parallel Pruning-Quantization)进行比较与分析的实验结果表明,QTD能够在相同压缩范围下实现更好的分类准确率。
文摘预训练模型通过自监督学习表示在非平行语料语音转换(VC)取得了重大突破。随着自监督预训练表示(SSPR)的广泛使用,预训练模型提取的特征中被证实包含更多的内容信息。提出一种基于SSPR同时结合矢量量化(VQ)和联结时序分类(CTC)的VC模型。将预训练模型提取的SSPR作为端到端模型的输入,用于提高单次语音转换质量。如何有效地解耦内容表示和说话人表示成为语音转换中的关键问题。使用SSPR作为初步的内容信息,采用VQ从语音中解耦内容和说话人表示。然而,仅使用VQ只能将内容信息离散化,很难将纯粹的内容表示从语音中分离出来,为了进一步消除内容信息中说话人的不变信息,提出CTC损失指导内容编码器。CTC不仅作为辅助网络加快模型收敛,同时其额外的文本监督可以与VQ联合优化,实现性能互补,学习纯内容表示。说话人表示采用风格嵌入学习,2种表示作为系统的输入进行语音转换。在开源的CMU数据集和VCTK语料库对所提的方法进行评估,实验结果表明,该方法在客观上的梅尔倒谱失真(MCD)达到8.896 d B,在主观上的语音自然度平均意见分数(MOS)和说话人相似度MOS分别为3.29和3.22,均优于基线模型,此方法在语音转换的质量和说话人相似度上能够获得最佳性能。
