Video retrieval is regarded as one of the most important in multimedia research. There are different types of representations for video i. e. low level representations and high level representations having different features based on the particular video annotating semantic concepts which appear in video shots is a challenging task. Integrated semantic based approach is proposed for similarity computations to improve the retrieval effectiveness in content based video retrieval. Video retrieval can be used for multiuser systems for video search and browsing which are useful in web applications. Here a DCT may be used that gives best performance among the different videos. We can introduce in this paper an interactive multiuser video retrieval framework composed by 1. Service oriented architecture-system which provides services for video semantic and semantic annotation, search and browsing for different domains 2. Web based interfaces for interactive query composition achieves browsing, annotation and visualization 3. A multitouch interface featuring gives collaborative natural interaction application. The performance of this technique is done with the help of cross over points of average precision and recall values of the data query videos fired on the whole database. The higher performance is achieved by HSV color space rather using RGB color space.

Retrieval techniques Color spaces Similarity measure Web based interface