中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

9 records (took 0.01 seconds)

    EVENT RECOGNITION USING IMAGE AND LOCATION INFORMATION
    1.
    发明申请
    EVENT RECOGNITION USING IMAGE AND LOCATION INFORMATION 审中-公开
    使用图像和位置信息的事件识别

    公开(公告)号：WO2010053513A1

    公开(公告)日：2010-05-14

    申请号：PCT/US2009/005832

    申请日：2009-10-27

    Applicant: EASTMAN KODAK COMPANY ,  YU, Jie ,  JOSHI, Dhiraj ,  LUO, Jiebo ,  HAO, Wei

    Inventor： YU, Jie ,  JOSHI, Dhiraj ,  LUO, Jiebo ,  HAO, Wei

    IPC: G06K9/00 ,  G06K9/46 ,  G06F17/30

    CPC classification number: G06K9/0063 ,  G06F17/30265 ,  G06K9/00664 ,  G06K9/4671

    Abstract: A method of recognizing an event depicted in an image from the image and a location information associated with the image is disclosed. The method includes acquiring the image and its associated location information; using the location information to acquire an aerial image(s) correlated to the location information; identifying the event using both the image and the acquired aerial image(s); and storing the event in association with the image for subsequent use.

    Abstract translation: 公开了一种从图像中识别图像中描绘的事件的方法和与图像相关联的位置信息。 该方法包括获取图像及其相关联的位置信息; 使用所述位置信息来获取与所述位置信息相关的空中图像; 使用图像和所获取的空间图像识别事件; 并将事件与图像相关联地存储以供后续使用。

    FINDING IMAGE CAPTURE DATE OF HARDCOPY MEDIUM
    2.
    发明申请
    FINDING IMAGE CAPTURE DATE OF HARDCOPY MEDIUM 审中-公开
    查找图像捕获日期

    公开(公告)号：WO2009151531A3

    公开(公告)日：2010-02-11

    申请号：PCT/US2009003098

    申请日：2009-05-19

    Applicant: EASTMAN KODAK CO ,  GALLAGHER ANDREW CHARLES ,  JOSHI DHIRAJ ,  LAWTHER JOEL SHERWOOD

    Inventor： GALLAGHER ANDREW CHARLES ,  JOSHI DHIRAJ ,  LAWTHER JOEL SHERWOOD

    IPC: H04N1/21 ,  G06K9/00 ,  H04N1/32

    CPC classification number: G06K9/3208 ,  G06K9/00221 ,  G06K2009/00322 ,  G06K2209/01 ,  G06K2209/27 ,  H04N2201/3205 ,  H04N2201/3214

    Abstract: A method of determining the image capture date of a scanned hardcopy medium having an image side and a non-image side, includes scanning the hardcopy medium to produce a scanned digital image; detecting handwritten annotations in the scanned digital image of the hardcopy medium; and using the handwritten annotations to determine the image capture date of the hardcopy medium by analyzing the handwritten annotations to identify names of people and associated ages; providing the names and lifespan information for a set of persons likely to appear in the hardcopy medium; and using the identified names of people and the associated ages along with the lifespan information to determine the image capture date.

    Abstract translation: 确定具有图像侧和非图像侧的扫描硬拷贝介质的图像捕获日期的方法包括扫描硬拷贝介质以产生扫描的数字图像; 检测硬拷贝介质的扫描数字图像中的手写注释; 并使用手写注释通过分析手写注释来识别人物名称和相关年龄来确定硬拷贝媒体的图像捕获日期; 为可能出现在硬拷贝媒体中的一组人提供姓名和生命周期信息; 并使用识别的人员名称和相关年龄以及生命周期信息来确定图像捕获日期。

    FINDING IMAGE CAPTURE DATE OF HARDCOPY MEDIUM
    3.
    发明申请
    FINDING IMAGE CAPTURE DATE OF HARDCOPY MEDIUM 审中-公开
    查找图像捕获日期

    公开(公告)号：WO2009151531A2

    公开(公告)日：2009-12-17

    申请号：PCT/US2009/003098

    申请日：2009-05-19

    Applicant: EASTMAN KODAK COMPANY ,  GALLAGHER, Andrew, Charles ,  JOSHI, Dhiraj ,  LAWTHER, Joel, Sherwood

    Inventor： GALLAGHER, Andrew, Charles ,  JOSHI, Dhiraj ,  LAWTHER, Joel, Sherwood

    IPC: H04N1/00

    CPC classification number: G06K9/3208 ,  G06K9/00221 ,  G06K2009/00322 ,  G06K2209/01 ,  G06K2209/27 ,  H04N2201/3205 ,  H04N2201/3214

    Abstract: A method of determining the image capture date of a scanned hardcopy medium having an image side and a non-image side, includes scanning the hardcopy medium to produce a scanned digital image; detecting handwritten annotations in the scanned digital image of the hardcopy medium; and using the handwritten annotations to determine the image capture date of the hardcopy medium by analyzing the handwritten annotations to identify names of people and associated ages; providing the names and lifespan information for a set of persons likely to appear in the hardcopy medium; and using the identified names of people and the associated ages along with the lifespan information to determine the image capture date.

    Abstract translation: 确定具有图像侧和非图像侧的扫描的硬拷贝介质的图像捕获日期的方法包括扫描硬拷贝介质以产生扫描的数字图像; 检测硬拷贝介质的扫描数字图像中的手写注释; 并使用手写注释通过分析手写注释来识别人物名称和相关年龄来确定硬拷贝媒体的图像捕获日期; 为可能出现在硬拷贝媒体中的一组人提供姓名和生命周期信息; 并使用识别的人员名称和相关年龄以及生命周期信息来确定图像捕获日期。

    AUTOMATED IMAGE ANNOTATION BASED UPON META-LEARNING OVER TIME
    5.
    发明申请
    AUTOMATED IMAGE ANNOTATION BASED UPON META-LEARNING OVER TIME 审中-公开
    基于元学习的自动图像注释随时间推移

    公开(公告)号：WO2009039480A2

    公开(公告)日：2009-03-26

    申请号：PCT/US2008/077196

    申请日：2008-09-22

    Applicant: THE PENN STATE RESEARCH FOUNDATION ,  DATTA, Ritendra ,  JOSHI, Dhiraj ,  LI, Jia ,  WANG, James, Z.

    Inventor： DATTA, Ritendra ,  JOSHI, Dhiraj ,  LI, Jia ,  WANG, James, Z.

    IPC: G06F17/21 ,  G06F17/00

    CPC classification number: G06F17/30265 ,  G06K9/6263

    Abstract: A principled, probabilistic approach to meta-learning acts as a go-between for a 'black- box' image annotation system and its users. Inspired by inductive transfer, the approach harnesses available information, including the black-box model's performance, the image representations, and a semantic lexicon ontology. Being computationally 'lightweight.' the meta-learner efficiently re-trains over time, to improve and/or adapt to changes. The black-box annotation model is not required to be re-trained, allowing computationally intensive algorithms to be used. Both batch and online annotation settings are accommodated. A "tagging over time" approach produces progressively better annotation, significantly outperforming the black-box as well as the static form of the meta-learner, on real-world data.

    Abstract translation: 一个有原则的，概率性的元学习方法可以作为“黑箱”图像注释系统及其用户的中介。 受到归纳转移的启发，该方法利用可用信息，包括黑盒模型的性能，图像表示和语义词典本体。 计算“轻量级”。 元学习者有效地重新训练，以改善和/或适应变化。 黑盒注释模型不需要重新训练，从而可以使用计算密集型算法。 批处理和在线注释设置都可以使用。 “随时间标记” 方法产生逐步更好的注释，显着优于黑盒以及元学习者的静态形式，以及真实世界的数据。

    METHOD FOR MEDIA BROWSING AND RELIVING
    6.
    发明申请
    METHOD FOR MEDIA BROWSING AND RELIVING 审中-公开
    媒体浏览和相关方法

    公开(公告)号：WO2012115829A1

    公开(公告)日：2012-08-30

    申请号：PCT/US2012/025177

    申请日：2012-02-15

    Applicant: EASTMAN KODAK COMPANY ,  LUO, Jiebo ,  JOSHI, Dhiraj ,  STUBLER, Peter, O. ,  DAS, Madirakshi ,  LEI, Phoury ,  SINGH, Vivek, Kumar

    Inventor： LUO, Jiebo ,  JOSHI, Dhiraj ,  STUBLER, Peter, O. ,  DAS, Madirakshi ,  LEI, Phoury ,  SINGH, Vivek, Kumar

    IPC: G06K9/00 ,  G06F3/041 ,  G11B27/00 ,  G06F3/048

    CPC classification number: G11B27/00 ,  G06F17/30044 ,  G06F17/30047

    Abstract: A method for viewing a collection of images or videos, includes analyzing the collection to determine properties of the images or videos and using the determined properties to produce icons corresponding to such properties; providing a time-varying display of the images or videos in the collection following an ordering of the images or videos in the collection and at least one of the corresponding icons; receiving a user selection of an icon; and changing the display of the images or videos in the collection following a reordering of the images or videos in the collection in response to the user selection.

    Abstract translation: 一种用于查看图像或视频的集合的方法，包括分析所述集合以确定所述图像或视频的属性，并使用所确定的属性来产生与所述属性对应的图标; 在所述集合中的图像或视频的排序和所述对应图标中的至少一个之后提供所述集合中的所述图像或视频的时变显示; 接收用户对图标的选择; 以及响应于用户选择，在收集中的图像或视频的重新排序之后，更改集合中的图像或视频的显示。

    RECOMMENDING USER IMAGE TO SOCIAL NETWORK GROUPS
    7.
    发明申请
    RECOMMENDING USER IMAGE TO SOCIAL NETWORK GROUPS 审中-公开
    推荐用户图像到社会网络组

    公开(公告)号：WO2011097041A2

    公开(公告)日：2011-08-11

    申请号：PCT/US2011/020063

    申请日：2011-01-04

    Applicant: EASTMAN KODAK COMPANY ,  YU, Jie ,  JOSHI, Dhiraj ,  LUO, Jiebo

    Inventor： YU, Jie ,  JOSHI, Dhiraj ,  LUO, Jiebo

    IPC: G06K9/62 ,  G06K9/46

    CPC classification number: G06K9/6218 ,  G06K9/00677 ,  G06K2209/27

    Abstract: A method of recommending social group(s) for sharing one or more user images, includes using a processor for acquiring the one or more user images and their associated metadata; acquiring one or more group images from the social group(s) and their associated metadata; computing visual features for the user images and the group images; and recommending social group(s) for the one of more user images using both the visual features and the metadata.

    Abstract translation: 一种推荐用于共享一个或多个用户图像的社团的方法包括使用处理器来获取一个或多个用户图像及其相关联的元数据; 从社会群体及其相关联的元数据获取一个或多个群组图像; 计算用户图像和组图像的视觉特征; 并使用视觉特征和元数据来为更多用户图像之一推荐社交群组。

    AUTOMATED IMAGE ANNOTATION BASED UPON META-LEARNING OVER TIME
    9.
    发明申请
    AUTOMATED IMAGE ANNOTATION BASED UPON META-LEARNING OVER TIME 审中-公开
    自动化图像估计基于元时间学习

    公开(公告)号：WO2009039480A3

    公开(公告)日：2009-05-22

    申请号：PCT/US2008077196

    申请日：2008-09-22

    Applicant: PENN STATE RES FOUND ,  DATTA RITENDRA ,  JOSHI DHIRAJ ,  LI JIA ,  WANG JAMES Z

    Inventor： DATTA RITENDRA ,  JOSHI DHIRAJ ,  LI JIA ,  WANG JAMES Z

    IPC: G06F17/21 ,  G06F17/00

    CPC classification number: G06F17/30265 ,  G06K9/6263

    Abstract: A principled, probabilistic approach to meta-learning acts as a go-between for a 'black- box' image annotation system and its users. Inspired by inductive transfer, the approach harnesses available information, including the black-box model's performance, the image representations, and a semantic lexicon ontology. Being computationally 'lightweight.' the meta-learner efficiently re-trains over time, to improve and/or adapt to changes. The black-box annotation model is not required to be re-trained, allowing computationally intensive algorithms to be used. Both batch and online annotation settings are accommodated. A "tagging over time" approach produces progressively better annotation, significantly outperforming the black-box as well as the static form of the meta-learner, on real-world data.

    Abstract translation: 元学习的原则性，概率性方法作为“黑盒子”图像注释系统及其用户的一个中介。 灵感来自感性传递，该方法利用了可用的信息，包括黑箱模型的性能，图像表示和语义词典本体。 在计算上“轻量级”。 元学习者随着时间的推移有效地重新训练，以改善和/或适应变化。 黑箱注释模型不需要重新训练，允许使用计算密集型算法。 批量和在线注释设置都可以收录。 随着时间的推移，“标记”方法可以逐渐更好的注释，显着优于实体数据的黑盒子和元学习者的静态形式。

Patent Agency Ranking