公开(公告)号：US08712785B2

公开(公告)日：2014-04-29

申请号：US13618414

申请日：2012-09-14

Applicant: Shuwu Wu ,  John Mantegna ,  Keren Perlmutter

Inventor： Shuwu Wu ,  John Mantegna ,  Keren Perlmutter

IPC: G10L19/022

CPC classification number: G10L19/00 ,  G10L19/0212 ,  G10L19/022 ,  G10L19/028 ,  G10L19/038 ,  G10L2019/0012

Abstract: A method and system for reduction of quantization-induced block-discontinuities arising from lossy compression and decompression of continuous signals, especially audio signals. One embodiment encompasses a general purpose, ultra-low latency, efficient audio codec algorithm. More particularly, the invention includes a method and apparatus for compression and decompression of audio signals using a novel boundary analysis and synthesis framework to substantially reduce quantization-induced frame or block discontinuity; a novel adaptive cosine packet transform (ACPT) as the transform of choice to effectively capture the input audio characteristics; a signal-residue classifier to separate the strong signal clusters from the noise and weak signal components (collectively called residue); an adaptive sparse vector quantization (ASVQ) algorithm for signal components; a stochastic noise model for the residue; and an associated rate control algorithm. The invention further includes corresponding computer program implementations of these and other algorithms.

Abstract translation: 用于减少由连续信号，特别是音频信号的有损压缩和解压缩引起的量化引起的块不连续性的方法和系统。 一个实施例包括通用的，超低延迟的高效音频编解码算法。 更具体地说，本发明包括一种使用新颖的边界分析和合成框架来压缩和解压缩音频信号的方法和装置，以大幅度减少量化引起的帧或块不连续性; 一种新颖的自适应余弦分组变换（ACPT）作为选择的有效捕获输入音频特性的变换; 信号残差分类器将强信号簇与噪声和弱信号分量（统称为残差）分离开来; 用于信号分量的自适应稀疏矢量量化（ASVQ）算法; 残留物的随机噪声模型; 和相关联的速率控制算法。 本发明还包括这些和其他算法的相应的计算机程序实现。

公开(公告)号：US20130173271A1

公开(公告)日：2013-07-04

申请号：US13618339

申请日：2012-09-14

Applicant: Shuwu Wu ,  John Mantegna ,  Keren Perlmutter

Inventor： Shuwu Wu ,  John Mantegna ,  Keren Perlmutter

IPC: G10L19/00

CPC classification number: G10L19/00 ,  G10L19/0212 ,  G10L19/022 ,  G10L19/028 ,  G10L19/038 ,  G10L2019/0012

Abstract: A method and system for reduction of quantization-induced block-discontinuities arising from lossy compression and decompression of continuous signals, especially audio signals. One embodiment encompasses a general purpose, ultra-low latency, efficient audio codec algorithm. More particularly, the invention includes a method and apparatus for compression and decompression of audio signals using a novel boundary analysis and synthesis framework to substantially reduce quantization-induced frame or block discontinuity; a novel adaptive cosine packet transform (ACPT) as the transform of choice to effectively capture the input audio characteristics; a signal-residue classifier to separate the strong signal clusters from the noise and weak signal components (collectively called residue); an adaptive sparse vector quantization (ASVQ) algorithm for signal components; a stochastic noise model for the residue; and an associated rate control algorithm. The invention further includes corresponding computer program implementations of these and other algorithms.

公开(公告)号：US20110282677A1

公开(公告)日：2011-11-17

申请号：US13191496

申请日：2011-07-27

Applicant: Shuwu WU ,  John Mantegna ,  Keren Perlmutter

Inventor： Shuwu WU ,  John Mantegna ,  Keren Perlmutter

IPC: G10L19/00

CPC classification number: G10L19/00 ,  G10L19/0212 ,  G10L19/022 ,  G10L19/028 ,  G10L19/038 ,  G10L2019/0012

Abstract: A method and system for reduction of quantization-induced block-discontinuities arising from lossy compression and decompression of continuous signals, especially audio signals. One embodiment encompasses a general purpose, ultra-low latency, efficient audio codec algorithm. More particularly, the invention includes a method and apparatus for compression and decompression of audio signals using a novel boundary analysis and synthesis framework to substantially reduce quantization-induced frame or block discontinuity; a novel adaptive cosine packet transform (ACPT) as the transform of choice to effectively capture the input audio characteristics; a signal-residue classifier to separate the strong signal clusters from the noise and weak signal components (collectively called residue); an adaptive sparse vector quantization (ASVQ) algorithm for signal components; a stochastic noise model for the residue; and an associated rate control algorithm. The invention further includes corresponding computer program implementations of these and other algorithms.

Abstract translation: 用于减少由连续信号，特别是音频信号的有损压缩和解压缩引起的量化引起的块不连续性的方法和系统。 一个实施例包括通用的，超低延迟的高效音频编解码算法。 更具体地说，本发明包括一种使用新颖的边界分析和合成框架来压缩和解压缩音频信号的方法和装置，以大幅度减少量化引起的帧或块不连续性; 一种新颖的自适应余弦分组变换（ACPT）作为选择的有效捕获输入音频特性的变换; 信号残差分类器将强信号簇与噪声和弱信号分量（统称为残差）分离开来; 用于信号分量的自适应稀疏矢量量化（ASVQ）算法; 残留物的随机噪声模型; 和相关联的速率控制算法。 本发明还包括这些和其他算法的相应的计算机程序实现。

公开(公告)号：US20090063164A1

公开(公告)日：2009-03-05

申请号：US12197645

申请日：2008-08-25

Applicant: Shuwu Wu ,  John Mantegna ,  Keren Perlmutter

Inventor： Shuwu Wu ,  John Mantegna ,  Keren Perlmutter

IPC: G10L19/00

CPC classification number: G10L19/00 ,  G10L19/0212 ,  G10L19/022 ,  G10L19/028 ,  G10L19/038 ,  G10L2019/0012

Abstract: A method and system for reduction of quantization-induced block-discontinuities arising from lossy compression and decompression of continuous signals, especially audio signals. One embodiment encompasses a general purpose, ultra-low latency, efficient audio codec algorithm. More particularly, the invention includes a method and apparatus for compression and decompression of audio signals using a novel boundary analysis and synthesis framework to substantially reduce quantization-induced frame or block discontinuity; a novel adaptive cosine packet transform (ACPT) as the transform of choice to effectively capture the input audio characteristics; a signal-residue classifier to separate the strong signal clusters from the noise and weak signal components (collectively called residue); an adaptive sparse vector quantization (ASVQ) algorithm for signal components; a stochastic noise model for the residue; and an associated rate control algorithm. The invention further includes corresponding computer program implementations of these and other algorithms.

Abstract translation: 用于减少由连续信号，特别是音频信号的有损压缩和解压缩引起的量化引起的块不连续性的方法和系统。 一个实施例包括通用的，超低延迟的高效音频编解码算法。 更具体地说，本发明包括一种使用新颖的边界分析和合成框架来压缩和解压缩音频信号的方法和装置，以大幅度减少量化引起的帧或块不连续性; 一种新颖的自适应余弦分组变换（ACPT）作为选择的有效捕获输入音频特性的变换; 信号残差分类器将强信号簇与噪声和弱信号分量（统称为残差）分离开来; 用于信号分量的自适应稀疏矢量量化（ASVQ）算法; 残留物的随机噪声模型; 和相关联的速率控制算法。 本发明还包括这些和其他算法的相应的计算机程序实现。

公开(公告)号：US20080089609A1

公开(公告)日：2008-04-17

申请号：US11868213

申请日：2007-10-05

Applicant: Keren Perlmutter ,  Sharon Perlmutter ,  Eric Wang ,  Paul Klamer ,  Leng Chua ,  Christopher Cookson ,  Massimiliano Gasparri

Inventor： Keren Perlmutter ,  Sharon Perlmutter ,  Eric Wang ,  Paul Klamer ,  Leng Chua ,  Christopher Cookson ,  Massimiliano Gasparri

IPC: G06K9/36 ,  G06K9/32

CPC classification number: H04N1/58 ,  G06K9/00758 ,  G06K9/32 ,  G06T3/00 ,  G06T3/403 ,  G06T7/32 ,  G06T2207/10024 ,  H04N5/262

Abstract: Images may be registered using temporal (time-based) and spatial information. In a film implementation, because film is a sequence of frames, using information from neighboring frames may enable a temporally smoother visual experience. In addition, it may be beneficial to take advantage of the fact that consecutive frames are often shifted similarly during the photographic process. Distortion measures may be used that discount candidate transformations that are considered to be too far from one or more preferred transformations, such as, for example, an optimal transformation from another frame or block or a currently-optimal transformation from the same frame/block. Composite color images may be processed to provide registration of underlying components.

Abstract translation: 可以使用时间（基于时间）和空间信息来注册图像。 在电影实现中，由于电影是一系列帧，使用来自相邻帧的信息可以使得能够在时间上更平滑的视觉体验。 另外，利用这样的事实可能是有益的，即在摄影过程中连续帧经常移动相似。 可以使用被认为与一个或多个优选变换太远的折扣候选变换的失真度量，例如来自另一帧或块的最佳变换或来自相同帧/块的当前最佳变换。 可以处理复合彩色图像以提供底层组件的注册。

公开(公告)号：US20070083364A1

公开(公告)日：2007-04-12

申请号：US11609081

申请日：2006-12-11

Applicant: Shuwu Wu ,  John Mantegna ,  Keren Perlmutter

Inventor： Shuwu Wu ,  John Mantegna ,  Keren Perlmutter

IPC: G10L19/00

CPC classification number: G10L19/00 ,  G10L19/0212 ,  G10L19/022 ,  G10L19/028 ,  G10L19/038 ,  G10L2019/0012

Abstract: A method and system for reduction of quantization-induced block-discontinuities arising from lossy compression and decompression of continuous signals, especially audio signals. One embodiment encompasses a general purpose, ultra-low latency, efficient audio codec algorithm. More particularly, the invention includes a method and apparatus for compression and decompression of audio signals using a novel boundary analysis and synthesis framework to substantially reduce quantization-induced frame or block-discontinuity; a novel adaptive cosine packet transform (ACPT) as the transform of choice to effectively capture the input audio characteristics; a signal-residue classifier to separate the strong signal clusters from the noise and weak signal components (collectively called residue); an adaptive sparse vector quantization (ASVQ) algorithm for signal components; a stochastic noise model for the residue; and an associated rate control algorithm. The invention further includes corresponding computer program implementations of these and other algorithms.

Abstract translation: 用于减少由连续信号，特别是音频信号的有损压缩和解压缩引起的量化引起的块不连续性的方法和系统。 一个实施例包括通用的，超低延迟的高效音频编解码算法。 更具体地说，本发明包括一种使用新颖的边界分析和合成框架来压缩和解压缩音频信号的方法和装置，以大幅减少量化引起的帧或块不连续性; 一种新颖的自适应余弦分组变换（ACPT）作为选择的有效捕获输入音频特性的变换; 信号残差分类器将强信号簇与噪声和弱信号分量（统称为残差）分离开来; 用于信号分量的自适应稀疏矢量量化（ASVQ）算法; 残留物的随机噪声模型; 和相关联的速率控制算法。 本发明还包括这些和其他算法的相应的计算机程序实现。

公开(公告)号：US20060067589A1

公开(公告)日：2006-03-30

申请号：US11236790

申请日：2005-09-28

Applicant: Keren Perlmutter ,  Sharon Perlmutter

Inventor： Keren Perlmutter ,  Sharon Perlmutter

IPC: G06K9/36

CPC classification number: G06T7/248 ,  G06K9/342 ,  G06K9/6202 ,  G06T5/005 ,  G06T5/50 ,  G06T7/0004 ,  G06T2207/10024

Abstract: Blotches may be identified and processed to reduce or eliminate the blotch. The blotch may be in just one of several separations and multiple separations may be used, for example, to identify the blotch. An implementation (i) compares a first component image of an image with a first component image of a reference image, (ii) compares a second component image of the image with a second component image of the reference image, and (iii) determines based on these comparisons whether the first component image of the image includes a blotch. Multiple image separations also, or alternatively, may be used, for example, to modify the blotch, as well as to evaluate whether a modification is beneficial.

Abstract translation: 可以识别和处理斑点以减少或消除斑点。 斑点可以仅在几个分离中的一个中，并且可以使用多个分离，例如以鉴定斑点。 （i）将图像的第一分量图像与参考图像的第一分量图像进行比较，（ii）将图像的第二分量图像与参考图像的第二分量图像进行比较，以及（iii）基于 在这些比较上，图像的第一分量图像是否包含斑点。 也可以使用多个图像分离，或者可以使用例如修饰斑点，以及评价修饰是否有益。

公开(公告)号：US20110085710A1

公开(公告)日：2011-04-14

申请号：US12849134

申请日：2010-08-03

Applicant: Keren Perlmutter ,  Sharon M. Perlmutter ,  Joshua Alspector ,  Mark Everingham ,  Alex Holub ,  Andrew Zisserman ,  Pietro Perona

Inventor： Keren Perlmutter ,  Sharon M. Perlmutter ,  Joshua Alspector ,  Mark Everingham ,  Alex Holub ,  Andrew Zisserman ,  Pietro Perona

IPC: G06K9/00

CPC classification number: G06K9/00288 ,  G06F17/30256 ,  G06K9/00281 ,  G06K9/6261 ,  G06K9/6263 ,  G11B27/105 ,  G11B27/28

Abstract: Images are searched to locate faces that are the same as a query face. Images that include a face that is the same as the query face may be presented to a user as search result images. Images also may be sorted by the faces included in the images and presented to the user as sorted search result images. The user may provide explicit or implicit feedback regarding the search result images. Additional feedback may be inferred regarding the search result images based on the user-provided feedback, and the results may be updated based on the user-provided and inferred feedback.

Abstract translation: 搜索图像以查找与查询面相同的面。 包括与查询面相同的面部的图像可以作为搜索结果图像呈现给用户。 图像也可以通过包括在图像中的面部进行排序，并且作为排序的搜索结果图像呈现给用户。 用户可以提供关于搜索结果图像的显式或隐含的反馈。 可以基于用户提供的反馈来推测关于搜索结果图像的附加反馈，并且可以基于用户提供的和推断的反馈来更新结果。

公开(公告)号：US20080056614A1

公开(公告)日：2008-03-06

申请号：US11856541

申请日：2007-09-17

Applicant: Keren Perlmutter ,  Sharon Perlmutter ,  Eric Wang ,  Paul Klamer ,  Christopher Cookson ,  Massimiliano Gasparri

Inventor： Keren Perlmutter ,  Sharon Perlmutter ,  Eric Wang ,  Paul Klamer ,  Christopher Cookson ,  Massimiliano Gasparri

IPC: G06K9/32

CPC classification number: G06T7/38 ,  G06K9/32 ,  G06T3/403 ,  G06T3/4084 ,  G06T5/003 ,  G06T5/10 ,  G06T5/50 ,  G06T7/12 ,  G06T7/13 ,  G06T2207/10016 ,  G06T2207/10024 ,  G06T2207/20064 ,  G06T2207/20192 ,  H04N1/58 ,  H04N5/262 ,  H04N9/093 ,  H04N9/11

Abstract: Separations or images relating to film or other fields may be registered using a variety of features, such as, for example: (1) correcting one or more film distortions; (2) automatically determining a transformation to reduce a film distortion; (3) applying multiple criteria of merit to a set of features to determine a set of features to use in determining a transformation; (4) determining transformations for areas in an image or a separation in a radial order; (5) comparing areas in images or separations by weighting feature pixels differently than non-feature pixels; (6) determining distortion values for transformations by applying a partial distortion measure and/or using a spiral search configuration; (7) determining transformations by using different sets of features to determine corresponding transformation parameters in an iterative manner; and (8) applying a feathering technique to neighboring areas within an image or separation.

Abstract translation: 可以使用各种特征来记录与胶片或其他领域有关的分离或图像，例如：（1）校正一个或多个胶片失真; （2）自动确定变换以减少胶片失真; （3）将多个品质标准应用于一组特征以确定用于确定变换的一组特征; （4）确定图像中的区域的转换或以径向顺序的分离; （5）通过与非特征像素不同地加权特征像素来比较图像或分离中的区域; （6）通过应用部分失真测量和/或使用螺旋搜索配置来确定变换的失真值; （7）通过使用不同的特征集确定变换，以迭代方式确定相应的变换参数; 和（8）对图像或分离中的相邻区域应用羽化技术。

公开(公告)号：US06885993B2

公开(公告)日：2005-04-26

申请号：US10061310

申请日：2002-02-04

Applicant: Shuwu Wu ,  John Mantegna ,  Keren Perlmutter

Inventor： Shuwu Wu ,  John Mantegna ,  Keren Perlmutter

IPC: G10L19/02 ,  G10L19/08

CPC classification number: G10L19/00 ,  G10L19/0212 ,  G10L19/022 ,  G10L19/028 ,  G10L19/038 ,  G10L2019/0012

Abstract: Compressing the digitized time-domain continuous input signal typically includes formatting the input signal into a plurality of time-domain blocks having boundaries, forming an overlapping time-domain block by prepending a fraction of a previous time-domain block to a current time-domain block, transforming each overlapping time-domain block to a transform domain block including a plurality of coefficients, partitioning the coefficients of each transform domain block into signal coefficients and residue coefficients, quantizing the signal coefficients for each transformed domain block and generating signal quantization indices indicative of such quantization, modeling the residue coefficients for each transform domain block as stochastic noise and generating residue quantization indices indicative of such quantization, and formatting the signal quantization indices and the residue quantization indices for each transform domain block as an output bit-stream. The continuous data may include audio data.