Dissolve Detection Based Shot Identification Using Singular Value Decomposition

Today, with the availability of large amount of video files, searching for videos with desired content is becoming a tedious task. The viewers require better control over the video data and for this reason the video browsing and indexing applications are being developed. These applications are required to segment the video into shots in the initial step. Shot boundary detection for shots separated by abrupt changes has been successful to a large extent, but detecting shot boundaries with gradual transitions in between, has been very challenging. In this paper, we propose a dual stage divide-and-merge approach to detect shots joined by dissolve type transitions. In the first stage we declare the video frames to be of either dissolve type or non-dissolve type. Later, we iteratively combine the sequences of non-dissolve frames that constitute the different shots. For experimental purpose, we have used the videos downloaded from the free video sharing website YouTube. The experimental results prove that the proposed technique is indeed effective.