公开(公告)号：US20120278077A1

公开(公告)日：2012-11-01

申请号：US13546153

申请日：2012-07-11

Applicant: Cha Zhang ,  Paul A. Viola ,  Pei Yin ,  Ross G. Cutler ,  Xinding Sun ,  Yong Rui

Inventor： Cha Zhang ,  Paul A. Viola ,  Pei Yin ,  Ross G. Cutler ,  Xinding Sun ,  Yong Rui

IPC: G10L17/00

CPC classification number: G06K9/6256 ,  G06K9/4614 ,  G10L25/78 ,  G10L2021/02166 ,  H04N7/147 ,  H04N7/15 ,  H04N21/42203 ,  H04N21/4223 ,  H04N21/4394 ,  H04N21/44008 ,  H04N21/44213 ,  H04N21/4788

Abstract: Systems and methods for detecting people or speakers in an automated fashion are disclosed. A pool of features including more than one type of input (like audio input and video input) may be identified and used with a learning algorithm to generate a classifier that identifies people or speakers. The resulting classifier may be evaluated to detect people or speakers.

Abstract translation: 公开了以自动方式检测人或扬声器的系统和方法。 可以识别包括多于一种类型的输入（例如音频输入和视频输入）的功能池，并与学习算法一起使用以生成识别人或扬声器的分类器。 可以评估所得分类器以检测人或扬声器。

公开(公告)号：US08234113B2

公开(公告)日：2012-07-31

申请号：US13221640

申请日：2011-08-30

Applicant: Cha Zhang ,  Paul A. Viola ,  Pei Yin ,  Ross G. Cutler ,  Xinding Sun ,  Yong Rui

Inventor： Cha Zhang ,  Paul A. Viola ,  Pei Yin ,  Ross G. Cutler ,  Xinding Sun ,  Yong Rui

IPC: G10L15/00

CPC classification number: G06K9/6256 ,  G06K9/4614 ,  G10L25/78 ,  G10L2021/02166 ,  H04N7/147 ,  H04N7/15 ,  H04N21/42203 ,  H04N21/4223 ,  H04N21/4394 ,  H04N21/44008 ,  H04N21/44213 ,  H04N21/4788

Abstract: Systems and methods for detecting people or speakers in an automated fashion are disclosed. A pool of features including more than one type of input (like audio input and video input) may be identified and used with a learning algorithm to generate a classifier that identifies people or speakers. The resulting classifier may be evaluated to detect people or speakers.

Abstract translation: 公开了以自动方式检测人或扬声器的系统和方法。 可以识别包括多于一种类型的输入（例如音频输入和视频输入）的功能池，并与学习算法一起使用以生成识别人或扬声器的分类器。 可以评估所得分类器以检测人或扬声器。

公开(公告)号：US20120141968A1

公开(公告)日：2012-06-07

申请号：US12961533

申请日：2010-12-07

Applicant: Na Li ,  Kai Yan Yang ,  Fanguang Kong ,  Yong Rui

Inventor： Na Li ,  Kai Yan Yang ,  Fanguang Kong ,  Yong Rui

IPC: G09B3/00

CPC classification number: G09B7/04

Abstract: Discussion evaluation may be provided. First, an assignment page including an evaluation link may be displayed and a user initiated input corresponding to the evaluation link may be received. Next, an evaluation view may be displayed in response to the received user initiated input. The displayed evaluation view may comprise an evaluation assistant data section and a raw discussion data section. Evaluation data may then be received in response to the displayed evaluation view.

Abstract translation: 可以提供讨论评估。 首先，可以显示包括评估链接的分配页面，并且可以接收与评估链接相对应的用户发起的输入。 接下来，可以响应于接收的用户启动的输入来显示评估视图。 显示的评估视图可以包括评估辅助数据部分和原始讨论数据部分。 响应于所显示的评估视图，可以接收评估数据。

公开(公告)号：US08068436B2

公开(公告)日：2011-11-29

申请号：US10686160

申请日：2003-10-15

Applicant: Yong Rui ,  Andres Vega-Garcia

Inventor： Yong Rui ,  Andres Vega-Garcia

IPC: H04L12/26

CPC classification number: H04L43/0882 ,  H04L43/022 ,  H04L43/045 ,  H04L43/0852

Abstract: Estimation of available bandwidth on a network uses packet pairs and spatially filtering. Packet pairs are transmitted over the network. The dispersion of the packet pairs is used to generate samples of the available bandwidth, which are then classified into bins to generate a histogram. The bins can have uniform bin widths, and the histogram data can be aged so that older samples are given less weight in the estimation. The histogram data is then spatially filtered. Kernel density algorithms can be used to spatially filter the histogram data. The network available bandwidth is estimated using the spatially filtered histogram data. Alternatively, the spatially filtered histogram data can be temporally filtered before the available bandwidth is estimated.

Abstract translation: 网络上可用带宽的估计使用数据包对和空间过滤。 分组对通过网络传输。 分组对的分散被用于生成可用带宽的样本，然后将其分类为分组以生成直方图。 箱体可以具有统一的箱体宽度，并且直方图数据可以老化，以便在估计中给予较小的重量。 然后将直方图数据进行空间滤波。 内核密度算法可用于对直方图数据进行空间过滤。 使用空间滤波的直方图数据估计网络可用带宽。 或者，空间滤波的直方图数据可以在估计可用带宽之前进行时间滤波。

公开(公告)号：US08024189B2

公开(公告)日：2011-09-20

申请号：US11425967

申请日：2006-06-22

Applicant: Cha Zhang ,  Paul A. Viola ,  Pei Yin ,  Ross G. Cutler ,  Xinding Sun ,  Yong Rui

Inventor： Cha Zhang ,  Paul A. Viola ,  Pei Yin ,  Ross G. Cutler ,  Xinding Sun ,  Yong Rui

IPC: G10L15/00

CPC classification number: G06K9/6256 ,  G06K9/4614 ,  G10L25/78 ,  G10L2021/02166 ,  H04N7/147 ,  H04N7/15 ,  H04N21/42203 ,  H04N21/4223 ,  H04N21/4394 ,  H04N21/44008 ,  H04N21/44213 ,  H04N21/4788

Abstract: Systems and methods for detecting people or speakers in an automated fashion are disclosed. A pool of features including more than one type of input (like audio input and video input) may be identified and used with a learning algorithm to generate a classifier that identifies people or speakers. The resulting classifier may be evaluated to detect people or speakers.

Abstract translation: 公开了以自动方式检测人或扬声器的系统和方法。 可以识别包括多于一种类型的输入（例如音频输入和视频输入）的功能池，并与学习算法一起使用以生成识别人或扬声器的分类器。 可以评估所得分类器以检测人或扬声器。

公开(公告)号：US20100325536A1

公开(公告)日：2010-12-23

申请号：US12487659

申请日：2009-06-19

Applicant: Fanguang Kong ,  Ying Xu ,  Shu Chen ,  Yong Rui

Inventor： Fanguang Kong ,  Ying Xu ,  Shu Chen ,  Yong Rui

IPC: G06F17/00

CPC classification number: G06F16/93

Abstract: A method for orchestrating various applications is described herein. A request to store a context information regarding a document may be received. An application in which the document is modified may be determined. The context information may be requested from the application. The context information may be stored. A request to recall the context information may be received. The context information may be displayed on a computer screen.

Abstract translation: 本文描述了用于编排各种应用的方法。 可以接收存储关于文档的上下文信息的请求。 可以确定文档被修改的应用程序。 可以从应用程序请求上下文信息。 可以存储上下文信息。 可以接收到召回上下文信息的请求。 上下文信息可以显示在计算机屏幕上。

公开(公告)号：US07739109B2

公开(公告)日：2010-06-15

申请号：US11035115

申请日：2005-01-12

Applicant: Yong Rui

Inventor： Yong Rui

IPC: G10L21/02 ,  H04B15/00 ,  H04M1/20

CPC classification number: H04N7/15 ,  H04N7/147 ,  H04N21/4143 ,  H04N21/42203 ,  H04N21/4223 ,  H04N21/4396 ,  H04N21/4788 ,  H04N21/6125

Abstract: A system and process for muting the audio transmission from a location of a participant engaged in a multi-party, computer network-based teleconference when that participant is working on a keyboard, is presented. The audio is muted as it is assumed the participant is doing something other than actively participation in the meeting when typing on the keyboard. If left un-muted the sound of typing would distract the other participant in the teleconference.

Abstract translation: 提出了一种系统和过程，用于在参与者正在使用键盘时从参与多方计算机网络的电话会议中的参与者的位置静音音频传输。 音频被静音，因为假设在键盘上键入时，参与者正在做积极参与会议的其他事情。 如果没有静音，打字的声音会分散电话会议中的其他参与者的注意力。

公开(公告)号：US20090290802A1

公开(公告)日：2009-11-26

申请号：US12125057

申请日：2008-05-22

Applicant: Xian-Sheng Hua ,  Guo-Jun Qi ,  Yong Rui ,  Tao Mei ,  Hong-Jiang Zhang

Inventor： Xian-Sheng Hua ,  Guo-Jun Qi ,  Yong Rui ,  Tao Mei ,  Hong-Jiang Zhang

IPC: G06K9/62

CPC classification number: G06K9/34

Abstract: The concurrent multiple instance learning technique described encodes the inter-dependency between instances (e.g. regions in an image) in order to predict a label for a future instance, and, if desired the label for an image determined from the label of these instances. The technique, in one embodiment, uses a concurrent tensor to model the semantic linkage between instances in a set of images. Based on the concurrent tensor, rank-1 supersymmetric non-negative tensor factorization (SNTF) can be applied to estimate the probability of each instance being relevant to a target category. In one embodiment, the technique formulates the label prediction processes in a regularization framework, which avoids overfitting, and significantly improves a learning machine's generalization capability, similar to that in SVMs. The technique, in one embodiment, uses Reproducing Kernel Hilbert Space (RKHS) to extend predicted labels to the whole feature space based on the generalized representer theorem.

Abstract translation: 所描述的并发多实例学习技术编码实例（例如，图像中的区域）之间的相互依赖性，以便预测将来实例的标签，以及如果需要，从这些实例的标签确定的图像的标签。 在一个实施例中，该技术使用并发张量来对一组图像中的实例之间的语义联系进行建模。 基于并发张量，可以应用秩1超对称非负张量因子分解（SNTF）来估计每个实例与目标类别相关的概率。 在一个实施例中，该技术在正则化框架中制定标签预测过程，其避免过拟合，并且显着地提高学习机器的泛化能力，类似于SVM中的标准预测过程。 在一个实施例中，该技术使用再生核希尔伯特空间（RKHS）来基于广义代表定理将预测标签扩展到整个特征空间。

公开(公告)号：US07589760B2

公开(公告)日：2009-09-15

申请号：US11286651

申请日：2005-11-23

Applicant: Cha Zhang ,  Bin Yu ,  Yong Rui

Inventor： Cha Zhang ,  Bin Yu ,  Yong Rui

IPC: H04N5/232

CPC classification number: H04N7/181 ,  H04N21/4223 ,  H04N21/4316

Abstract: A computer network-based distributed presentation system and process is presented that controls the display of one or more video streams output by multiple video cameras located across multiple presentation sites on display screens located at each presentation site. The distributed presentation system and process provides the ability for a user at a site to customize the screen configuration (i.e., what video streams are display at any one time and in what format) for that site via a two-layer display director module. In the design layer of the module, a user interface is provided for a user to specify display priorities dictating what video streams are to be displayed on the screen over time. These display priorities are then provided to the execution layer of the module which translates them into probabilistic timed automata and uses the automata to control what is displayed on the display screen.

Abstract translation: 提出了一种基于计算机网络的分布式呈现系统和过程，其控制由位于每个呈现站点的显示屏幕上的多个呈现站点上的多个摄像机输出的一个或多个视频流的显示。 分布式呈现系统和过程提供了一个站点用户通过两层显示导演模块定制屏幕配置（即，任何一个时间和以什么格式显示什么视频流）的能力。 在模块的设计层中，为用户提供用户界面，以指定显示优先级，指定在屏幕上随时间显示哪些视频流。 然后将这些显示优先级提供给模块的执行层，将其转换为概率定时自动机，并使用自动机来控制显示屏上显示的内容。

公开(公告)号：US20090125461A1

公开(公告)日：2009-05-14

申请号：US11958050

申请日：2007-12-17

Applicant: Guo-Jun Qi ,  Xian-Sheng Hua ,  Yong Rui ,  Hong-Jiang Zhang ,  Shipeng Li

Inventor： Guo-Jun Qi ,  Xian-Sheng Hua ,  Yong Rui ,  Hong-Jiang Zhang ,  Shipeng Li

IPC: G06F15/18

CPC classification number: G06N99/005

Abstract: Multi-label active learning may entail training a classifier with a set of training samples having multiple labels per sample. In an example embodiment, a method includes accepting a set of training samples, with the set of training samples having multiple respective samples that are each respectively associated with multiple labels. The set of training samples is analyzed to select a sample-label pair responsive to at least one error parameter. The selected sample-label pair is then submitted to an oracle for labeling.

Abstract translation: 多标签主动学习可能需要对分类器训练一组具有每个样本的多个标签的训练样本。 在示例实施例中，一种方法包括接受一组训练样本，其中该组训练样本具有多个相应样本，每个样本分别与多个标签相关联。 分析该组训练样本以响应于至少一个误差参数来选择样本标签对。 然后将选定的样品标签对提交给oracle进行标记。