The integration of MIR modalities

This last chapter begins by treating problematic aspects and the inefficiency of data processing algorithms, such as the specific cases of noise and loss of information . Then the functionality and the solutions of the technical - organic integration of the content-based operational modes of TR, VR, VDR and AR are discussed. It is stressed that in order to efficiently achieve MIR, the integration between content-based modes must be organically linked to the integration with the semantic modes of treatment and research of information and documents. This is discussed together with the issue of integrating MIR with other management systems in use for other types of documents, non-multimedia or non-digital that together constitute the heritage of a given library or archive. Each theoretical and paradigmatic perspective in this sense is valid only if it can be put into the development of real multimedia search and retrieval interfaces, enabling the end user to set and conduct a search by combining various textual, visual and sound modes. In all this, there is always a negative response to the hypothesis that only content-based treatment can be the most effective way to search for information, explaining the importance that the descriptive and semantic metadata still have in treating all types of document. However much they have been developed, the elaborate algorithms that can calculate the possible interpretations of a set of objective contentual information are unable to bridge the so called semantic gap , which continues to be created between low-level mechanical consideration of an object’s appearance in face of the high-level evaluation of the human idea of that object.