Abstract: A new algorithm based on the co-histogram between two consecutive frames is presented. In the algorithm, the similarity is defined based on the co-histogram symmetry which is high in the same shot, but low in a shot transition. The similarity effectively reveals the transition characteristics. The difference of similarities is used to enhance the abrupt transition features, and a Finite-State Automata is used to improve the robustness in detecting gradual transition. Experiments testing various videos show that the algorithm achieves a performance with 95.6％ recall and 95.8％ precision.

Key words: video retrieval, shot boundary detection, co-histogram, abrupt transition, gradual transition