Rule-based scene boundary detection for semantic video segmentation

In this paper, we present a technique for semantic video segmentation. Our technique uses a combination of low-level features and high-level rules which organizes the video into scenes, shots and key-frames. Features in color domain is calculated and utilized for detecting the key-frames and estimating the similarity between shots. By applying a set of high-level rules, similar shots are merged and the scene boundaries are determined. Finally, a likelihood function is designed for improving the accuracy of scene boundary results. Experimental results from several Hollywood movies have demonstrated and show a better performance of both precision and recall has been achieved comparing with other existing works.