HIERARCHICAL IMAGE MODELING FOR OBJECT-BASED MEDIA RETRIEVAL

Images an more structurally complex than the data stored in the tradit ional DBMSs. Many systems perform media retrieval based on image featu re matching; the fact that images have structures is ignored. On the o ther hand, we view images as compound objects containing many componen t objects. Each component object corresponds to a region(s) that is a visually and semantically meaningful entity (e.g. car, man, etc.). As images/objects truly have hierarchical structures and have both visual and semantics properties, we argue that image retrieval using either semantics and image matching alone is insufficient. In this paper, we introduce a hierarchical structure for image modeling to support image retrieval using combinations of semantic expressions and visual examp les at both the whole image and object levels. We give formal definiti ons of a multimedia query language and a system implementation based o n this image modeling. We also address associated query processing iss ues and discuss the portability and extendibility of our approach.