中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

7 records (took 0.025 seconds)

    METHOD AND APPARATUS FOR AUTOMATICALLY SUMMARIZING VIDEO
    2.
    发明申请
    METHOD AND APPARATUS FOR AUTOMATICALLY SUMMARIZING VIDEO 有权
    用于自动总结视频的方法和装置

    公开(公告)号：US20140161351A1

    公开(公告)日：2014-06-12

    申请号：US14183070

    申请日：2014-02-18

    Applicant: Google Inc.

    Inventor： Jay N. Yagnik

    IPC: G06K9/00

    CPC classification number: G06K9/00536 ,  G06F17/30787 ,  G06F17/30802 ,  G06F17/30843 ,  G06K9/00711

    Abstract: One embodiment of the present invention provides a system that automatically produces a summary of a video. During operation, the system partitions the video into scenes and then determines similarities between the scenes. Next, the system selects representative scenes from the video based on the determined similarities, and combines the selected scenes to produce the summary for the video.

    Abstract translation: 本发明的一个实施例提供一种自动产生视频摘要的系统。 在操作期间，系统将视频划分为场景，然后确定场景之间的相似性。 接下来，系统基于确定的相似度从视频中选择代表性场景，并且组合所选择的场景以产生视频的摘要。

    Techniques for utilizing and adapting a prediction model
    4.
    发明授权
    Techniques for utilizing and adapting a prediction model 有权
    利用和适应预测模型的技术

    公开(公告)号：US09122986B2

    公开(公告)日：2015-09-01

    申请号：US13669073

    申请日：2012-11-05

    Applicant: Google Inc.

    Inventor： Jay N. Yagnik

    IPC: G06N5/04 ,  G06N5/02

    CPC classification number: G06N5/02 ,  G06F17/30035 ,  G06N5/04 ,  G06N5/046

    Abstract: A computer-implemented technique of providing relevant search results to a user of a website at a query time. The technique can include receiving, at a computing device having one or more processors, a query from the user, the query corresponding to a description of potential search results desired by the user. The technique can further include retrieving a user history corresponding to previous user interactions with the website and determining a context of the user corresponding to an interaction of the user with the website at the query time. The relevant search results can be determined based on the query, the user history, and the context of the user and a prediction model, and be provided to the user via updating of a webpage presented to the user. The technique can further include adapting the prediction model based on a prediction event and set of corresponding prediction event features.

    Abstract translation: 一种在查询时间向网站用户提供相关搜索结果的计算机实现技术。 该技术可以包括在具有一个或多个处理器的计算设备处接收来自用户的查询，该查询对应于用户期望的潜在搜索结果的描述。 该技术可以进一步包括检索与先前用户与网站的交互相对应的用户历史，并且在查询时间确定与用户与网站的交互相对应的用户的上下文。 可以基于用户的查询，用户历史和上下文以及预测模型来确定相关搜索结果，并且通过更新呈现给用户的网页来向用户提供相关的搜索结果。 该技术还可以包括基于预测事件和相应的预测事件特征的集合来适应预测模型。

    TECHNIQUES FOR UTILIZING AND ADAPTING A PREDICTION MODEL
    5.
    发明申请
    TECHNIQUES FOR UTILIZING AND ADAPTING A PREDICTION MODEL 有权
    利用和适应预测模型的技术

    公开(公告)号：US20150154493A1

    公开(公告)日：2015-06-04

    申请号：US13669073

    申请日：2012-11-05

    Applicant: Google Inc.

    Inventor： Jay N. Yagnik

    IPC: G06N5/02

    CPC classification number: G06N5/02 ,  G06F17/30035 ,  G06N5/04 ,  G06N5/046

    Abstract: A computer-implemented technique of providing relevant search results to a user of a website at a query time. The technique can include receiving, at a computing device having one or more processors, a query from the user, the query corresponding to a description of potential search results desired by the user. The technique can further include retrieving a user history corresponding to previous user interactions with the website and determining a context of the user corresponding to an interaction of the user with the website at the query time. The relevant search results can be determined based on the query, the user history, and the context of the user and a prediction model, and be provided to the user via updating of a webpage presented to the user. The technique can further include adapting the prediction model based on a prediction event and set of corresponding prediction event features.

    Abstract translation: 一种在查询时间向网站用户提供相关搜索结果的计算机实现技术。 该技术可以包括在具有一个或多个处理器的计算设备处接收来自用户的查询，该查询对应于用户期望的潜在搜索结果的描述。 该技术可以进一步包括检索与先前用户与网站的交互相对应的用户历史，并且在查询时间确定与用户与网站的交互相对应的用户的上下文。 可以基于用户的查询，用户历史和上下文以及预测模型来确定相关搜索结果，并且通过更新呈现给用户的网页来向用户提供相关的搜索结果。 该技术还可以包括基于预测事件和相应的预测事件特征的集合来适应预测模型。

    Principal component analysis based seed generation for clustering analysis
    6.
    发明授权
    Principal component analysis based seed generation for clustering analysis 有权
    基于主成分分析的种子生成用于聚类分析

    公开(公告)号：US08660370B1

    公开(公告)日：2014-02-25

    申请号：US13755373

    申请日：2013-01-31

    Applicant: Google Inc.

    Inventor： Sangho Yoon ,  Jay N. Yagnik ,  Mei Han ,  Vivek Kwatra

    IPC: G06K9/62

    CPC classification number: G06K9/6247 ,  G06K9/6223

    Abstract: Clustering algorithms such as k-means clustering algorithm are used in applications that process entities with spatial and/or temporal characteristics, for example, media objects representing audio, video, or graphical data. Feature vectors representing characteristics of the entities are partitioned using clustering methods that produce results sensitive to an initial set of cluster seeds. The set of initial cluster seeds is generated using principal component analysis of either the complete feature vector set or a subset thereof. The feature vector set is divided into a desired number of initial clusters and a seed determined from each initial cluster.

    Abstract translation: 诸如k均值聚类算法的聚类算法被用于处理具有空间和/或时间特征的实体的应用中，例如表示音频，视频或图形数据的媒体对象。 使用产生对初始集群种子集合敏感的结果的聚类方法对代表实体特征的特征向量进行分区。 使用完整特征向量集或其子集的主成分分析来生成初始簇种子集合。 特征向量集合被分为期望数量的初始簇和从每个初始簇确定的种子。

Patent Agency Ranking