Knowledge-assisted retrieval of video data
Young Francis Day, Rolan Paul, Abdul Ghafoor, Ashfaq Khokhar · OSTI OAI (U.S. Department of Energy Office of Scientific and Technical Information) · 1996
Several levels of spatio-temporal knowledge are needed to facilitate the retrieval of video sequences. The first level is the domain knowledge of the physical objects which are of the interest of the applications, presented in the video. The second level of knowledge defines spatial/temporal events identified from video sequences. The third level of knowledge provides the semantic relationships among the spatial/temporal events. In this paper, we present an approach to address the latter two issues. We use a set of n-ary operators as the basic constructors for the construction of knowledge about spatial/temporal events. To facilitate query processing, knowledge about the relationships among spatial/temporal events are organized using a hierarchical object-oriented abstraction.