An Effective Approach on CBVR Based on High Level Semantics

All videos will eventually become fully digital-there seems to be no way around it. Consequently, digital video databases will become more and more pervasive and finding video in large digital video databases will become a problem just like it is a problem today to find video in analog video databases. This poses a major challenge of video annotation and finding out a suitable video for a particular application.

In this paper, the problem of automatic video annotation is presented. This means associating semantic meaning with video segments which aids in Content-Based Video Retrieval (CBVR). A novel framework of structural video analysis is presented in this paper, which focuses on the processing of low-level visual data cues to obtain high-level (structural and semantic) video interpretations using Artificial Neural Networks (ANN). It is observed that integrated feature vector gives the best of the results as compared to any single feature vector considered alone.