A Framework for Semantic Modeling of Video Data for Content-Based Indexing and Retrieval

In this paper, we propose a semantic-based abstraction of video data for capturing the spatio-temporal concepts associated with various objects in an input image or in a sequence of video frames. This abstraction can manifest itself eeectively in conceptualizing events and views in multimedia data as perceived by individual users. The objective is to provide an eecient mechanism for handling content-based queries, with the minimum amount of processing performed on raw data during query evaluation. We introduce a multi-level architecture for video data and semantics are shared among various levels. This leads to a multi-level indexing/searching mechanism. At the nest level of granularity, video data can be indexed based on simple contents such as identifying appearance of objects and faces. For management of information at higher levels of abstractions, an object-oriented paradigm is proposed which provides a comprehensive video database architecture.