|
Domestic video clips usually contain Chinese captions. The captions include rich semantic information and they are valuable descriptions of the video content. Extracting the high-level semantic of caption can help people to understand the content of the video easily and it also can make the video retrieval more efficiently. In this paper, an approach, which is based on semantic correlation between sentences in a caption text, is proposed to generate the summarization of the caption. Firstly we get the caption text of the video and segment the caption text into words and sentences to complete information segmentation. Secondly using the clustering algorithm, sentences are clustered to implement information combination. Finally representative sentences are extracted from each class to generate the summarization of caption, which can be considered to be video summarization Experimental results indicate that the method is feasible and useful. |
|
Keywords:Video Retrieval, Caption, Automatic Summarization, Chinese Word Segmentation |
|