Indexing and archiving of multimedia contents

Indexing is not traditionally considered an engineering subject and is not typically aimed at an audience trained in the natural sciences, a field in which engineers are well-versed. As a library and information science practice, indexing is primarily a content-based task, where interpretation and judgment are key to representing a document in a way that makes it retrievable.

Indeed, indexing is the technique of describing a document’s content and the information it contains in such a way that it can be retrieved when searches are conducted on any of the topics it covers. The challenge, therefore, lies in knowing how to characterize and represent the information in a document so that it can be easily linked to topics of interest. By linking a query to represented and synthesized content, indexing allows users to navigate the vast sea of documents and organize their knowledge. Indexing thus falls under what has been referred to in recent years as intellectual techniques [1][2][3] .

Indexing is a long-standing need but a relatively recent practice that has been the focus of the emerging disciplines of information science and library science. Developed over the past two centuries (in the 20th century in the case of information science), indexing has primarily concerned textual content, and it is on this type of content that it has refined its processes, developed its concepts, and honed its methods.

The purpose of this article is to examine how indexing evolves when, on the one hand, it is applied to computerized or electronic documents, and on the other hand, when it involves non-textual content such as videos or audio. Although indexing of non-textual documents has long been practiced for analog media such as photographs, films, drawings, and other diagrams, the introduction of digital technologies has profoundly transformed indexing, turning it into “information retrieval” and paving the way for a new paradigm: “fine-grained content indexing” or “content-based indexing.”

As a result, indexing today is a subject of interest to both engineers and information specialists or librarians. Whether approached from the perspective of computer science and mathematics or based on documentary and interpretive principles, indexing can take on very different forms. The purpose of this article is not to repeat what is already known and published on the topic of information retrieval. Rather, its aim is to highlight the challenges of documentary indexing when dealing with non-textual electronic content, in order...