News Video Story Segmentation Based on Topic Caption Text and Audio Information

This paper presents a novel scheme of news video story segmentation with the combination of topic caption frame and audio information. First of all, a new method of topic caption frame detection is proposed in which the topic caption frame is detected by features extraction from frame differences, the topic caption lasting time and times of caption transition in the same shot. Afterwards, the silence clip between two continuous news story boundaries is detected according to short-time average energy and short-time average zero-crossing rate. Moreover, one news video is segmented into a series of news story clips on the basis of topic caption text and silence clip. The experimental results showed that the proposed method could effectively detect topic caption text and news story clip boundaries for news video.