|
In this thesis, we introduce a "Multimedia Object Composition
Model"(MOCM). This model includes a script language
specification that is used to describe the spatial and temporal
composition relationships among multimedia objects. This model
also has a preprocessor to translate users' description into
host language. The preprocessor parses the scripts and
generates a spatial and temporal relation table of all objects.
Then through this table, preprocessor can generate an
executable program. For multimedia application designers, they
can use the tools provided by this system with host programming
language ( ex. C, C++ ) to write multimedia application systems
easily. For example, we can develope multimedia presentation
systems, multimedia explanations, CAI and so on. The MOCM's
media object class hierarchy we developed is an open ended
architecture. It hides the implementation details from users.
Currently, we only support three types of media : audio, video,
and text. If a user want to include another type of MOCM's
media object, he only have to inherit suitable super class
given in the frameworks to create a new class, and just call
the library functions associated with new media.
|