Abstract:This paper seeks to provide a brief overview of those developments which have taken the theory and practice of image and video retrieval into the digital age. Drawing on a voluminous literature, the context in which visual information retrieval takes place is followed by a consideration of the conceptual and practical challenges posed by the representation and recovery of visual material on the basis of its semantic content. An historical account of research endeavours in content-based retrieval, directed towa… Show more
“…techniques specifically for image retrieval as a technology which can be applied to multimedia IR. Enser (2008a) relates the failure of CBIR systems based on A.I. techniques to fulfil their promise, with critics in the information science community demonstrating through experimentation that users do not find low level features useful for search.…”
Section: Retrievalmentioning
confidence: 99%
“…Commercial systems that proposed using such an approach have failed to make any headway as a result. A focus on the more semantic aspects of the content (Enser, 2008a) has proved to be problematic also as some concepts in an image are intrinsic to it and are not physically present (e.g. a picture of a politician involved in an election -the 'politician' is identifiable, but the concept of an 'election' is more difficult to detect).…”
Various kinds of knowledge organisation (such as thesauri) are routinely used to label or tag multimedia content such as images and music, to support information retrieval i.e. user search for such content. In this paper we outline why this is the case, in particular focusing on the semantic gap between content and concept based multimedia retrieval. We survey some indexing vocabularies used for multimedia retrieval, and argue that techniques such as thesauri will be needed for the foreseeable future in order to support users in their need for multimedia content. In particular we argue that Artificial Intelligence (A.I.) techniques are not mature enough to solve the problem of indexing multimedia conceptually, and will not be able to replace human indexers for the foreseeable future. ACKNOWLEDGEMENTS I am very grateful to my colleague Deborah Lee for her advice and links to/on thesauri for music and video, and also to David Bawden in confirming the lack of work in those domains. Thanks also go to Stella Dextre Clarke and Judi Vernau for their very constructive comments on various drafts of the paper.
“…techniques specifically for image retrieval as a technology which can be applied to multimedia IR. Enser (2008a) relates the failure of CBIR systems based on A.I. techniques to fulfil their promise, with critics in the information science community demonstrating through experimentation that users do not find low level features useful for search.…”
Section: Retrievalmentioning
confidence: 99%
“…Commercial systems that proposed using such an approach have failed to make any headway as a result. A focus on the more semantic aspects of the content (Enser, 2008a) has proved to be problematic also as some concepts in an image are intrinsic to it and are not physically present (e.g. a picture of a politician involved in an election -the 'politician' is identifiable, but the concept of an 'election' is more difficult to detect).…”
Various kinds of knowledge organisation (such as thesauri) are routinely used to label or tag multimedia content such as images and music, to support information retrieval i.e. user search for such content. In this paper we outline why this is the case, in particular focusing on the semantic gap between content and concept based multimedia retrieval. We survey some indexing vocabularies used for multimedia retrieval, and argue that techniques such as thesauri will be needed for the foreseeable future in order to support users in their need for multimedia content. In particular we argue that Artificial Intelligence (A.I.) techniques are not mature enough to solve the problem of indexing multimedia conceptually, and will not be able to replace human indexers for the foreseeable future. ACKNOWLEDGEMENTS I am very grateful to my colleague Deborah Lee for her advice and links to/on thesauri for music and video, and also to David Bawden in confirming the lack of work in those domains. Thanks also go to Stella Dextre Clarke and Judi Vernau for their very constructive comments on various drafts of the paper.
“…La recherche par le contenu est un champ de recherche scientifique très productif qui crée des méca-nismes pour récupérer de grandes collections d'images (Tsai, McGarry et Tait, 2006). Ce champ de recherche commence à se développer à la fin des années 1970, sous la forme de bases de données construites spécialement pour le stockage et le repérage d'images (Enser, 2008). Le paradigme de recherche et de repérage d'images par contenu fonctionne à partir du contenu explicite de l'image numérisée, c'est-à-dire à partir des pixels.…”
Section: La Recherche Et Le Repérage D'imagesunclassified
“…Le paradigme de recherche et de repérage d'images par contenu fonctionne à partir du contenu explicite de l'image numérisée, c'est-à-dire à partir des pixels. Certains auteurs se montrent sceptiques quant au choix de ce terme, puisqu'il peut prêter à confusion avec les aspects sémantiques du processus (Enser, 2008). Ce type de repérage utilise les métadonnées caractérisant les formes, les couleurs et les textures de l'image.…”
Section: La Recherche Et Le Repérage D'imagesunclassified
“…Ce type de recherche et de repérage d'images est discuté dans la littérature scientifique à partir du milieu des années 1990 (Enser, 2008). Une partie de la recherche doit se réaliser verbalement (Enser, 2000) ; on utilise les mots-clés attribués aux images ou bien on recherche en mode plein texte.…”
Section: La Recherche Et Le Repérage D'imagesunclassified
Cet article présente l’analyse de fiches descriptives de banques d’images et leur normalisation. Il présente tout d’abord quelques concepts théoriques sur les images et leur description, sur les modes de recherche et de repérage des images, sur les métadonnées internes (EXIF, IPTC et XMP) ainsi que sur la normalisation. L’analyse des éléments de la description formelle des photographies dans 30 banques d’images tient compte des éléments potentiellement utiles pour la recherche et le repérage, de la terminologie employée, de la longueur des légendes et de la fréquence d’utilisation des métadonnées internes. L’article conclut que les fiches descriptives ne sont pas normalisées (bien qu’il existe une tendance à réutiliser quelques éléments coïncidant avec les champs des métadonnées IPTC) et recommande d’employer des métadonnées internes pour normaliser la description formelle des images.This article analyzes the work sheets used for image or photographic data bases and the standards that are used. The theoretical concepts associated with images and their description, their search and retrieval, the metadata (EXIF, IPTC and XMP) as well as the standards used are also outlined. The analysis of the elements used in the formal description of photographs in 30 image data bases considered the elements that would be helpful during the search and retrieval process, the terminology used, the length of captions and the frequency of the use of internal metadata. Even though there appears to be a tendency to use some of the elements that are identical to those in the IPTC metadata fields, the article concludes that the work sheets are not standardized. The author recommends the use of internal metadata in order to help standardize the formal description of images.Este artículo presenta el análisis de una serie de fichas descriptivas de bancos de imágenes y su normalización. En primer lugar expone algunos conceptos teóricos en cuanto a las imágenes y su descripción, a los modos de búsqueda y localización de las imágenes, a los metadatos internos (EXIF, IPTC y XMP) y a la normalización. El análisis de los elementos que integran la descripción formal de las fotografías en 30 bancos de imágenes toma en cuenta elementos potencialmente útiles para la búsqueda y localización, la terminología empleada, la extensión de las leyendas y la frecuencia de utilización de los metadatos internos. El artículo concluye que las fichas descriptivas no están normalizadas (aunque exista una tendencia a reutilizar algunos elementos, que coinciden con los campos de los metadatos IPTC) y recomienda el empleo de metadatos internos para normalizar la descripción formal de las imágenes
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.