专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

WO2009067159A2 MEDIA ASSET EVALUATION BASED ON SOCIAL RELATIONSHIPS 审中-公开
标题翻译：基于社会关系的媒体资产评估
公开(公告)号：WO2009067159A2
公开(公告)日：2009-05-28
申请号：PCT/US2008/012709
申请日：2008-11-12
申请人： EASTMAN KODAK COMPANY , BIRKELUND, John Richard , WOOD, Mark D.
发明人： BIRKELUND, John Richard , WOOD, Mark D.
IPC分类号： G06F17/30
CPC分类号： G06F17/30265 , G06F17/30247 , G06K9/00664 , G06Q50/01 , G11B27/105 , G11B27/28 , H04N21/4756
摘要： A method for evaluating media assets in a set of media assets including the step of obtaining information defining at least social relationships of an audience. The social relationships include relationships between a member of the audience and someone recorded in at least one of the media assets. For each media asset in the set of media assets, the method includes identifying one or more people recorded in the digital media asset, computing, for each person recorded in the digital media asset, the person's expected social significance relative to the audience based at least upon the social relationships, and computing an overall measurement of the social significance of the digital media asset as a function of the expected social significance of each person recorded in the asset. In addition, the method includes storing the overall measurement for each digital media asset in a processor-accessible memory system.
摘要翻译：一种用于评估一组媒体资产中的媒体资产的方法，包括获取至少定义观众的社会关系的信息的步骤。社会关系包括观众成员与至少一个媒体资产中记录的人之间的关系。对于媒体资产集合中的每个媒体资产，该方法包括识别在数字媒体资产中记录的一个或多个人，针对记录在数字媒体资产中的每个人计算至少至少相对于观众的人的预期社会意义在社会关系上，计算数字媒体资产的社会意义的总体衡量，作为资产中记录的每个人的预期社会意义的函数。此外，该方法包括将每个数字媒体资产的总体测量存储在处理器可访问存储器系统中。

2. 发明申请

WO2010059188A3 METHOD FOR EVENT-BASED SEMANTIC CLASSIFICATION 审中-公开
公开(公告)号：WO2010059188A3
公开(公告)日：2010-05-27
申请号：PCT/US2009/006064
申请日：2009-11-12
申请人： EASTMAN KODAK COMPANY , DAS, Madirakshi , LOUI, Alexander C. , WOOD, Mark D.
发明人： DAS, Madirakshi , LOUI, Alexander C. , WOOD, Mark D.
IPC分类号： G06F17/30 , G06K9/46
摘要： A method of automatically classifying images in a consumer digital image collection, includes generating an event representation of the image collection; computing global time-based features for each event within the hierarchical event representation; computing content-based features for each image in an event within the hierarchical event representation; combining content-based features for each image in an event to generate event-level content-based features; and using time-based features and content-based features for each event to classify an event into one of a pre-determined set of semantic categories.

3. 发明公开

EP2347349A2 METHOD FOR EVENT-BASED SEMANTIC CLASSIFICATION 审中-公开
标题翻译： VERFAHREN ZUR SEMANTISCHEN KLASSIFIKATION AUF EREIGNISBASIS
公开(公告)号：EP2347349A2
公开(公告)日：2011-07-27
申请号：EP09761050.5
申请日：2009-11-12
申请人： Eastman Kodak Company
发明人： DAS, Madirakshi , LOUI, Alexander C. , WOOD, Mark D.
IPC分类号： G06F17/30 , G06K9/46
CPC分类号： G06F17/30598 , G06F17/30247 , G06F17/30265 , G06K9/00664
摘要： A method of automatically classifying images in a consumer digital image collection, includes generating an event representation of the image collection; computing global time-based features for each event within the hierarchical event representation; computing content-based features for each image in an event within the hierarchical event representation; combining content-based features for each image in an event to generate event-level content-based features; and using time-based features and content-based features for each event to classify an event into one of a pre-determined set of semantic categories.
摘要翻译：将在消费者数字图像集合中检测到的基于时间的事件分组为超级事件的方法。该方法包括：生成基于时间的事件边界; 计算事件间持续时间; 并将事件分组为基于事件间持续时间的称为超级事件的集群。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式