Hybrid query for video database system

Many video retrieval systems support text and low-level feature-based querying, however only single homogeneous queries are allowed. In this paper we present a Video Database System (VDBS) based on our video data model that captures the hierarchical structure and contents of video to support hybrid query. The system supports query using similarity-based matching of low-level visual features as well as exact matching of textual attributes. The experiment shows that hybrid query gives more accurate results when compared to single query using text or image alone by eliminate results that are similar in colour but has different semantic, or vice versa.