Resumen
The role of the emerging field of visual information retrieval (VIR) systems is to go beyond text-based descriptors to elicit, store and retrieve the "imagery-based" information content in visual media. The basic premise behind VIR systems is that images and video are first-class information-bearing entities and that users should be able to query their content as easily as they query textual documents, without necessarily using manual annotation. Most applications for VIR fall between automated pixel-oriented information models and fully human-assisted database schemes. Most of the current VIR systems are limited in the query types they can handle. Development of a comprehensive language for visual assets is a far more difficult task. However, query specification for visual informations should not be performed exclusively with an example or specification-based paradigm but through a collection of different tools. Unfortunately, much effort has not been directed to establish criteria for evaluating, benchmarking and comparing VIR systems.