机构地区: 人民大会堂
出 处: 《计算机工程与应用》 2005年第27期157-159,共3页
摘 要: 情节代表帧选取方法是视频语义分析和基于内容的视频检索的很重要的方法。代表帧的使用大大减少了视频索引的数据量,同时也为视频摘要和检索提供了一种快捷的方法。该文在子镜头的关键帧提取方法基础上,利用模糊C-均值聚类算法,实现了一种基于子镜头聚类的情节代表帧选取方法。实验证明该方法计算简单,可以较好地代表视频情节。 Selecting episode representation frame is one of the important processes in video semantic analysis and content-based video retrieval.Representation frame can dramatically reduce the data of the video indexing and provides a shortcut method for video abstraction and retrieval.An algorithm for selecting episode representation frames by using an approach of key frame extraction based on multiple characters and C-Mean fuzzy clustering is detailed in the paper. Experimental results show that the approach is computationally simple and able to adapt to the visual content.
关 键 词: 代表帧选取 子镜头聚类 情节 模糊 均值聚类算法
领 域: [自动化与计算机技术] [自动化与计算机技术]