@book {IOPORT.00856304,
    editor       = {McKevitt, Paul},
    title        = {Integration of natural language and vision processing (Volume II). Intelligent multimedia. Repr. from the journal Artificial Intelligence Review 9, Nos. 2-3 (1995).},
    year         = {1995},
    isbn         = {0-7923-3758-1},
    edition      = {Repr. from the journal Artificial Intelligence Review 9, Nos. 2-3 (1995)},
    pages        = {170~p.},
    publisher    = {Dordrecht: Kluwer Academic Publishers},
    abstract     = {[For part I see Zbl 0827.68088.] This compilation contains eleven articles and a book review of M. Maybury's ``Intelligent Multimedia Interfaces'', which have been reprinted from the ``Artificial Intelligence Review'', Volume 9, Nos. 2-3. Four of the eleven articles are site descriptions and discuss some recent developments in the field of intelligent multimedia, which are currently shaping in the labs of the DFKI (Germany), Apple Computer (USA), the University of Brighton (England) and LIMSI-CRNS (France). The work reported spans from Apple's authoring tools to the DFKI's WIP system for generation of multimedia documents from a formal description of the communicative intent of a planned presentation. Two papers are general overviews of intelligent multimedia and discuss multimedia interfaces. M. Maybury gives a general overview of current techniques for parsing and generating multiple media (e.g., text, graphics, maps, gestures) using multiple sensory modalities (e.g., auditory, visual, tactile). Systems are discussed that have integrated parsing and generation to enable multimedia dialogue in the context of intelligent interfaces. Fundamental problems which require further research are also outlined. O. Stock discusses a hypermedia based approach to human-computer interaction and the integration of natural language processing with other media such as gestures and graphics. He illustrates his work on multimodal user interfaces by describing the ALFRESCO and MAIA system. Two papers discuss multimedia presentation. E. Andre and T. Rist survey the WIP system for generating coherent presentation employing textual and visual material. A major issue of their work concerns the planning procedures and process coordination required for adequate content determination, medium selection and content realization in different media. Y. Arens and E. Hovy discuss the design of a multimedia interaction manager called Cicero, which performs run-time media coordination and allocation. Cicero adapts to the presentation context and knows what it is presenting in order to provide coherent extended human-machine dialogues. Three papers look at communication from the point of view of icons and animation. C. Beardon stresses the fact that the syntactic and semantic structure of natural language can be retained, while iconic images replace words. As an illustration, he describes the design principles behind a new iconic language, called IconText. Design considerations for a visual language, which may allow users with different languages to communicate freely using icons, are also considered by M. Yazdani and S. Mealing. They illustrate their approach by a system which enables hotel booking where there is an icon-based dialogue between a potential guest and a hotel manager. A. Narayanan et al. discuss principles of language visualization, i.e., mapping language expressions onto graphical forms, where the graphical forms convey the meaning of the language expressions. A universal visualization system architecture is proposed, with graphical primitives corresponding to those of Schank's Conceptual Dependency theory. They show an example story (based on a subset of the restaurant script) and how it can be mapped into visual primitives and animation sequences.},
    reviewer     = {U.Hahn (Freiburg i.Br.)},
    identifier   = {00856304},
}