Shot Boundary Detection and Keyframe Extraction Based on Scale Invariant Feature Transform

Abstract

In shot boundary detection, the key technology is to compute the visual content discontinuity values between consecutive video frames. In this paper, a unified framework is proposed to detect the shot boundaries and extract the keyframes of a shot. Firstly, the scale invariant feature transform (SIFT) is adopted to compute the visual content discontinuity values. Then a new method, which is called the local double threshold shot boundary detection (LDT-SBD), is used to detect shot boundaries. Lastly, two mechanisms are proposed to extract keyframe. Experimental results show the framework is effective and has a good performance.