Dance videos are interesting and semantics-intensive. At the same time, they are complex type of videos, when compared to all other types such as sports, news and movie videos. In fact, dance video is the one which is less explored by the researchers across the globe. This paper presents a dance video data model to represent the semantics of the dance videos with different granularity levels, identified by the components of the accompanying song. Secondly, the paper proposes a multi-level index structure to efficiently handle containment, temporal, spatial and spatiotemporal query types. Moreover, the index structure has been designed to consider the queries with different levels of constraints. Finally, the paper presents the experimental results depicting the performance of different types of dance video queries.
An Adaptive Index Structure for High-Dimensional Similarity Search