ABSTRACT
Most of today’s content-based multimedia analysis and retrieval systems tend to follow a low-level approach when tackling both content analysis and retrieval tasks, thus falling short of benefits uprising from higher-level interpretation and knowledge. The role of additional information in the sense of semantics, context, and implicit or explicit knowledge is gaining focus on the task of bridging the semantic and conceptual gap that exists between humans and computers, in order to further facilitate human-computer interaction and scene content understanding. This chapter focuses on modeling and exploiting contextual knowledge toward efficient multimedia content understanding. As discussed below, this type of information acts as a simulation of the human visual perception scheme, by taking into account all contextual information relative to the visual content of a scene [1]. As a result, the notion of context, provided that it will be properly modeled and justified, may be used to improve the performance of knowledge-assisted analysis, semantic indexing, and retrieval of multimedia content.
