公开(公告)号：US08983945B1

公开(公告)日：2015-03-17

申请号：US13342617

申请日：2012-01-03

申请人： Zhiyi Xie ,  Varun Kacholia ,  Minya Dai ,  Patrick Hung ,  Kun Zhang ,  Jonathan Eng ,  Nicola Muscettola ,  Omer Bar-or ,  Aparna Chennapragada ,  Nitin Khandelwal

发明人： Zhiyi Xie ,  Varun Kacholia ,  Minya Dai ,  Patrick Hung ,  Kun Zhang ,  Jonathan Eng ,  Nicola Muscettola ,  Omer Bar-or ,  Aparna Chennapragada ,  Nitin Khandelwal

IPC分类号： G06F17/30

CPC分类号： G06F17/3082 ,  G06F17/30823

摘要： The present disclosure relates to the identification of video content. In one aspect, a method includes generating a query based on bibliographic data. The method also includes obtaining a collection of resources responsive to the query, wherein one or more of the resources include text and video content. The method further includes calculating occurrence scores for the resources. A particular occurrence score for a particular resource is based at least in part on the bibliographic data matching text included in the particular resource and the text being associated with video content. The method further includes selecting one or more resources as including video content identified by the bibliographic data using the occurrence scores. The method further includes storing data associating the selected resources with the bibliographic data.

摘要翻译： 本公开涉及视频内容的识别。 一方面，一种方法包括基于书目数据生成查询。 该方法还包括响应于查询获得资源集合，其中一个或多个资源包括文本和视频内容。 该方法还包括计算资源的发生得分。 特定资源的特定出现分数至少部分地基于包括在特定资源中的书目数据匹配文本和与视频内容相关联的文本。 该方法还包括使用发生得分将一个或多个资源选择为包括由书目数据标识的视频内容。 该方法还包括存储将所选择的资源与书目数据相关联的数据。

公开(公告)号：US08953836B1

公开(公告)日：2015-02-10

申请号：US13363116

申请日：2012-01-31

申请人： Gheorghe Postelnicu ,  Richard Iles Tucker ,  Murali Krishna Viswanathan ,  Nicola Muscettola ,  Varun Kacholia

发明人： Gheorghe Postelnicu ,  Richard Iles Tucker ,  Murali Krishna Viswanathan ,  Nicola Muscettola ,  Varun Kacholia

IPC分类号： G06K9/00

CPC分类号： G06K9/00758

摘要： Systems and methods are provided herein relating to real-time duplicate detection of video content. Fingerprints can be generated for an uploaded video. The fingerprints can be used to match the uploaded video to a set of matching videos. The set of matching videos can be filtered based on the type of match, and the quality of the match. A unique cluster-id can be generated for the uploaded video containing an upload time, and that unique cluster-id can then be modified to associate the uploaded video with a cluster-id of potential duplicates. Cluster-ids can then be used in the context of a search to filter results that have identical cluster-ids. The benefits in using real-time duplicate detection can better maximize user experiences in a video sharing service that contains potential duplicates of the same content.

摘要翻译： 本文提供了与视频内容的实时重复检测有关的系统和方法。 可以为上传的视频生成指纹。 指纹可用于将上传的视频与一组匹配的视频相匹配。 匹配视频的集合可以根据匹配的类型和匹配的质量进行过滤。 可以为包含上传时间的上传视频生成唯一的cluster-id，然后可以修改唯一的cluster-id，将上传的视频与可能重复的集群ID相关联。 然后可以在搜索的上下文中使用Cluster-id来过滤具有相同集群ids的结果。 使用实时重复检测的好处可以更好地最大限度地最大限度地增加包含相同内容的潜在重复性的视频共享服务中的用户体验。