Scene break detection and classification using a block-wise difference method

We introduce a new approach to the detection and the classification of effects in video sequences. We deal with the effects such as cuts, fades, dissolves, and camera motions. A global motion compensation based on block matching and a measure of block mean intensities are used to detect all the effects. We determine the dominant motion vectors to detect the camera motions for each shot and observe the ratio of the difference intensity variation of blocks in consecutive frames to detect change effects between video shots. The approach can handle the complex motions during gradual effects as well as the precisely detection of effect lengths. Both synthetic and real evidences are presented to demonstrate how this approach can efficiently classify effects in video sequences involving significant motions.