...
首页> 外文期刊>Multimedia Tools and Applications >Semantic video database system with semi-automatic secondary-content generation capability
【24h】

Semantic video database system with semi-automatic secondary-content generation capability

机译:具有半自动辅助内容生成功能的语义视频数据库系统

获取原文
获取原文并翻译 | 示例
   

获取外文期刊封面封底 >>

       

摘要

A semantic video database system, based on an interactive approach that maps low-level features to high-level concepts, is proposed. A database of ontological semantic object models allows the user to get information about specific semantic objects, such as certain actors or other features of a TV drama. The system searches the database for key frames within the video to detect similarities in detailed or "low-level" features, such as the color, area, and position of a specific part of the frame. Since image recognition techniques are limited in their ability to fully identify and compare images, we propose an additional function in which a coarse model is used to recover a greater number of similar key frames, thus providing more relevant results. From these results, the content provider can select relevant key frames interactively; the matched objects in them are then automatically annotated according to descriptions that are added into the model by content provider. Therefore, more complex content can be generated with greater accuracy by using a combination of application-oriented operations. The system has high potential for use in object-based interactive multimedia applications. We also present an object-based video content generation application called the Drama Characters' Popularity Voting System.
机译:提出了一种基于交互式方法的语义视频数据库系统,该方法将低级特征映射到高级概念。本体语义对象模型的数据库允许用户获取有关特定语义对象的信息,例如某些演员或电视剧的其他特征。系统在数据库中搜索视频中的关键帧,以检测详细或“低级”功能的相似性,例如帧的特定部分的颜色,区域和位置。由于图像识别技术无法完全识别和比较图像,因此我们提出了一项附加功能,其中使用粗略模型来恢复更多数量的相似关键帧,从而提供更多相关结果。根据这些结果,内容提供者可以交互选择相关的关键帧。然后,根据内容提供者添加到模型中的描述,为其中的匹配对象自动添加注释。因此,通过结合使用面向应用程序的操作,可以更准确地生成更复杂的内容。该系统在基于对象的交互式多媒体应用中具有很高的潜力。我们还介绍了一个基于对象的视频内容生成应用程序,称为“戏剧人物的人气投票系统”。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号