The last two decades have resulted in a substantial progress in the multimedia and storage technology that has led to building of a large repository of digital image, video, and audio data. There are a number of text-search engines on the web and incidentally, the sites hosting them are amongst the busiest sites. However, searching for a multimedia content is not as easy because the multimedia data, as opposed to text, needs many stages of pre-processing to yield indices relevant for querying. Since an image or a video sequence can be interpreted in numerous ways, there is no commonly agreed-upon vocabulary. Thus, the strategy of manually assigning a set of labels to a multimedia data, storing it and matching the stored label with a query will not be effective. Besides, the large volume of video data makes any assignment of text labels a massively labor intensive effort.

In recent years research has focused on the use of internal features of images and videos computed in an automated or semi-automated way [1],[2]. Automated analysis calculates statistics which can be approximately correlated to the content features. This is useful as it provides information without costly human interaction. The common strategy for automatic indexing had been based on using syntactic features alone. However, due to its complexity of operation, there is a paradigm shift in the research of identifying semantic features [3]. Userfriendly Content-Based Retrieval (CBR) systems operating at semantic level would identify motion-features as the key besides other features like color, objects etc., because motion (either of camera motion or shot editing) adds to the meaning of the content. The focus of present motionbased systems had been mainly in identifying the princi-