Visual and textual content based indexing and retrieval

Abstract.

Although there are several research and commercial systems in content-based multimedia indexing and retrieval, more effort should be focused on the quality of retrieval, particularly regarding how well the retrieval results correspond to what the user really wants. In this context, we present an approach that may contribute to the quality of indexing and retrieval. It supports both textual and visual indexing and retrieval that obtains results with a higher degree of quality, and it incorporates domain knowledge into the index without any special efforts on the part of the user.