{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}


Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
A GENERIC ARCHITECTURE FOR AUTOMATED CONSTRUCTION OF MULTIMEDIA PRESENTATIONS FREDERIC BES, MURIEL JOURDAN AND FARID KHANTACHE Unité de Recherche INRIA Rhône-Alpes ZIRST, 655 avenue de l’Europe, Montbonnot, 38334 St Ismier CEDEX France E-mail: {Frederic.Bes,Muriel.Jourdan,Farid.Khantache}@inrialpes.fr We are interested in applications that automatically generate a dynamic multimedia presentation adapted to the needs of the user from a database of XML fragments of basic elements (video, images, paragraphs, etc.). In this paper we present an architecture which goal is to support developers of such applications. This architecture takes benefits from both the use of transformation sheets and the use of a constraint solver. Transformation sheets are used to provide different renderings that depend on discrete parameters; whereas the constraint solver aims to handle continuous parameters and to optimize both content selection and formatting instructions taking into account global criteria like the total duration of the generated multimedia document. We illustrate the use of this architecture by presenting an application that builds a personalized TV sport news automatically. 1 Introduction TV-like multimedia presentations will be more and more used to communicate information: training courses, medical reports, news, on-line shopping catalogues, etc. However, designing high-quality multimedia presentations is known to be a complex, time-consuming and error prone task [3,5,10] whatever is the authoring tool used. Moreover, the information overload problem associated with the Internet and the frequent updates of some information like news or medical reports raise a crucial need of information services that are customized for each individual user [4,15]. Typical examples of such information services are: educational courses adapted to the student’s knowledge, automatic design of medical reports, personalized TV-news. Whereas many works have been done in those areas on the run-time adaptation of the navigational structure of a hypermedia document [8], few works focus on the automatic generation of multimedia presentations. Moreover existing works often propose solutions linked to some particular area and it is not easy to evaluate how these specific experiments can help to implement other kind of applications. This is why we propose an architecture to help the development of personalized multimedia information services. Starting from the twofold analysis that using transformation sheets is today the best way to associate different presentations to the same set of objects depending on discrete parameters; and that global criteria, like a user preference on the total
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
duration of the multimedia document get at the end, are easy to handle by using optimization techniques, the architecture takes benefits from these two technologies.
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}