Media annotation - Archive ouverte HAL Accéder directement au contenu
Chapitre D'ouvrage Année : 2002

Media annotation

Résumé

Advanced multimedia systems require the development of complex processes and tools to enable the underlying structure and content to be easily understood and to facilitate access and manipulation of such information. Clearly defined semantics is an essential characteristic in this context. Modelling must provide, through the use of pattern recognition, indexing or classifying tools, high level descriptions or abstractions of multimedia data, content and structure. A query support must allow the user to query by the idea he has of their appearance rather than by their exact content. Querying by example (and counter-examples) or allowing for flexible queries seem to be natural in this case. Our proposal is a conceptual model which enables adaptable and reusable multimedia content. Particular emphasis has been put on detecting characteristic features of multimedia documents seen as semi-structured data and querying them, for instance with examples. The different uses of current query languages applied to the model are detailed. This chapter is organised into three sections. The first one deals with the generation of describers. The second one enhances the faceted description, according to the different dimensions of the media. The third part is dedicated to multimedia querying, the interpretation of which relies on describers.
Fichier non déposé

Dates et versions

hal-03783543 , version 1 (22-09-2022)

Identifiants

Citer

Claude Chrisment, Florence Sèdes. Media annotation. Djeraba, Chabane. Multimedia Mining: A Highway to Intelligent Multimedia Documents, 22 (Chapter 10), Kluwer Academic, pp.197-211, 2002, Multimedia Systems and Applications Series (MMSA), 1-4020-7247-3. ⟨10.1007/978-1-4615-1141-0_11⟩. ⟨hal-03783543⟩
10 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More