Efficient Video Summarization Based on Maximally Stable Extremal Regions

摘要：

In this paper,a novel framework is proposed in the video summarization application. Considering the large number of video data and calculation,an efficient structure with the name of Multi-Dimension Precedence Stable Adaptive Binary Tree (MDPSABT),which contains the video content information in all nodes,is presented. In order to analyze and match the video content more accurately,the popular Maximally Stable Extremal Regions (MSER) detector is applied,and then a descriptor,which is proved to be simple but appropriate in this video application,is used to represent the region features. A series of experiments show that the proposed framework performs with high efficiency and accuracy on challenging video data sets.

关键词： video summarization Maximally Stable Extremal Region(MSER) Multi-Dimension Precedence Stable Adaptive Binary Tree (MDPSA-BT) adaptive binary tree

作者: Chaochao Lu Tong Yu Xing Wang

作者单位: Software Institute,Nanjing University,Nanjing,P.R.C

会议类型: 国际会议

会议名称: 2011 International Conference on Opto-Electronics Engineering and Information Science(2011光电电子工程与信息科学国际会议 ICOEIS 2011)

会议地点: 西安

会议语种:英文

页码: 182-187

在线出版日期: 2011-12-23（万方平台首次上网日期，不代表论文的发表时间）