关键词: 镜头边界检测, 角点检测, 特征跟踪

Abstract: Partitioning a video sequence into shots is the first step toward video-content analysis and content-based video browsing and retrieval. A novel video shot boundary detection algorithm was presented based on the feature tracking. First, the proposed algorithm extracts a set of corner-points as features from the first frame of a shot. Then, based on the Kalman filtering, the extracted features are tracked with window-matching method from the succeed frames. According to the characteristic pattern of pixels value changing within the tracked windows, the measure of shot boundary detection can be obtained to confirm the types of transition and the time interval of each gradual transition. The experimental results illustrate that the proposed algorithm is effective and robust with low computational complexity.

Key words: Shot boundary detection, Corner detection, Feature tracking