No abstract

Existing key-frame extraction methods are basically video summary oriented; yet the index task of key-frames is ignored. This paper presents a novel key-frame extraction approach which can be available for both video summary and video index. First a dynamic distance separability algorithm is advanced to divide a shot into subshots based on semantic structure, and then appropriate key-frames are extracted in each subshot by SVD decomposition. Finally, three evaluation indicators are proposed to evaluate the performance of the new approach. Experimental results show that the proposed approach achieves good semantic structure for semantics-based video index and meanwhile produces video summary consistent with human perception.

Abstract. Shot boundary detection is the first important task of content-based video retrieval. In this paper, a new SBD algorithm is proposed aiming to obtain accurate detection, and its performances are evaluated with different types of video. This algorithm computes distance ratio between within-class and between-class of two group frames, rather than the difference between two frames, which can resist light effects and camera/object movements in the same shot. The experimental results show that this universal algorithm can gain higher precision and recall.

No abstract