Video shot detection using cumulative colour histogram

Abstract

The shot boundary detection is the fundamental step in video indexing and retrieval. In this paper a new method for video shot boundary detection based on slope and y-intercept parameters of the straight line fitted to the cumulative plot of color histogram is proposed. These feature vectors are extracted from every video frames and the frame dissimilarity values are compared against a threshold to identify the cuts and fades present in the video sequence. Experiments have been conducted on TRECVID video database to evaluate the effectiveness of the proposed model. A comparative analysis with other models is also provided to reveal the superiority of the proposed model for shot detection. © 2013 Springer.