中文摘要：

针对如何在语义层次上形成视频摘要问题，提出了一种基于支持向量机的风光纪录片语义视频摘要算法．利用支持向量机对镜头关键帧进行语义分类，对每类镜头关键帧根据引入的镜头“重要性”函数提取构建视频摘要的帧．改变重要性函数阈值的大小，可以方便地得到不同粒度的视频摘要．实验结果表明，该算法形成的视频摘要较好地表达了视频的内容．

英文摘要：

Constructing the video summarization in semantic level is very important. An algorithm of video summarization based on support vector machine （SVM） in semantic level for the natural scenery documentary is presented. The shot key frames are classified by SVM and the frames constructing the video summarization are selected from the shot key frames of every class by the importance function introduced. The scalable video summarization from coarse to fine level of detail can be achieved by changing the threshold of important function. Experiment results indicate that the proposed algorithm performs satisfactorily.