In this paper we introduce the idea of enhancing the audio presentation of a
multimedia museum guide by using the PDA screen to travel throughout a
fresco and identify the various details in it. During the presentation, a sequence
of pictures is synchronized with the audio commentary, and the transitions
among the pictures are planned according to cinematic techniques.
The theoretical background is presented, discussing the language of
cinematography and the Rhetorical Structure Theory to analyze dependency
relationships inside a text. In building the video clips, a set of strategies similar
to those used in documentaries was employed. Two broad classes of
strategies have been identified. The first class encompasses constraints
imposed by the grammar of cinematography, while the second deals with
conventions normally used in guiding camera movements in the production of
documentaries.
The results of a preliminary evaluation are also presented and discussed.