tvDBMS - A video database management system incorporating a thematic indexing model

This demonstration presents a novel video database system, tvDBMS, which caters for complex and long videos, such as documentaries, educational videos, etc. As compared to relatively structured format videos like CNN news or commercial advertisements, this database system has the capacity to work with long and unstructured videos. tvDBMS metadata contains information about segments (combinations of frames) and scenes (collections of segments that represent the same content). This metadata is organized in such a way that it can be used to navigate a theme (concept or idea). Annotations describing scenes are linked in hierarchical manner to create a story line in the video, where thematic indexing is used to develop a video catalogue. Thematic indexing is a novel way to track a story in a video. A video contains many themes, which are implicitly related to each other. In order to resolve queries about particular scenes in a video, the scenes are stored hierarchically, which provides 'is-a' or 'have-part' relations between them. We have tested our model on one of the hour-long documentaries made for television by the Earl Mountbatten of Burma. The results show that a user can easily query for natural scenes or events such as Japanese Soldiers landing in Burma in World War 2, Allied forces defending the border, air bombings, soldiers marching in fields etc. tvDBMS is divided into three basic components. The Composer, which provides a user with tools to generate metadata for videos, create thematic indices and database connectivity of metadata and video data, as shown in figure (1). The Query Processor deals with all the queries to be performed and how to track a particular story line for an event or an object, whereas the Query Output Window provides twelve best searched portions of videos of a query