基于子镜头聚类方法的关键帧提取技术

Method of Key Frame Extraction Based on Sub-Shot Clustering

  • 摘要: 分析主流的关键帧提取技术,提出了一种基于子镜头聚类的关键帧提取算法. 该方法在重新定位镜头的起始和终止帧号后,利用帧与帧之间的颜色直方图特征,通过子镜头检测和聚类进行关键帧提取. 实验结果表明,该算法具有良好的适应性,既降低了关键帧提取算法的计算复杂度,正确率高,同时能有效避免关键帧的冗余,达到了很好的关键帧提取效果.

     

    Abstract: By analyzing the known techniques of frame extraction, a new method of key frame extraction (KFE) is proposed based on sub-shot clustering in this paper. Using the features of color histogram between successive frames, the key frames could be extracted via sub-shot detection and clustering after the relocation of the shot boundary. Experimental results show that, the proposed method is adaptable, accurate, and effective. The improvement of key frame extraction lies in the fact that it not only reduces the complexity of extracting process, but also avoids redundancies of key frames effectively.

     

/

返回文章
返回
Baidu
map