Multimedia on the Semantic Web

First generation Web-content encodes information in handwritten (HTML) Web pages. Second generation Web content generates HTML pages on demand, e.g. by filling in templates with content retrieved dynamically from a database or transformation of structured documents using style sheets (e.g. XSLT). Third generation Web pages will make use of rich markup (e.g. XML) along with metadata (e.g. RDF) schemes to make the content not only machine readable but also machine processable --- a necessary pre-requisite to the Semantic Web.

While text-based content on the Web is already rapidly approaching the third generation, multimedia content is still trying to catch up with second generation techniques. Our research at CWI is focussing on tools to support second and third generation multimedia.

Second generation multimedia

Multimedia document processing has a number of fundamentally different requirements from text which make it more difficult to incorporate within the document processing chain. In particular, multimedia transformation uses different document and presentation abstractions, its formatting rules cannot be based on text-flow, it requires feedback from the formatting back-end and is hard to describe in the functional style of current style languages. We are currently developing the Cuypers multimedia presentation generation that address these issues.

Third generation multimedia

Adequately annotated multimedia is a key pre-requisite for this multimedia variant of the Semantic Web. Unfortunately, current multimedia authoring tools provide little support for producing annotated multimedia presentations. Much of the underlying semantics of the overall multimedia presentation and the media fragments it contains, remains implicit and is only present in the head of the author. In contrast, in the Cuypers system discussed above, it is relatively easy to generate such annotations automatically. Since the entire presentation-generation process in the Cuypers system is based on explicitly encoded knowledge, this knowledge can be preserved and encoded as rich metadata annotations in the final-form presentation. Note that such metadata annotations can arise from different knowledge sources and describe different abstraction levels. For example, when the system is used to generated richly annotated SMIL, the metadata section of the SMIL document may contain metadata about the individual media items (as retrieved from the underlying media database), the rhetorical structure of the overall presentation, domain-specific knowledge of the application, etc. It could also generate a report of the design rules and user profiles that were used to justify the chosen the end-result (e.g. the machine-readable equivalent of "this presentation contains much hi-end video because it is generated for users with a broadband network environment"). This could, for instance, be used by the browsers to help with automatic detection of errors in the settings of the end-user's profile.

$Id: index.html,v 1.4 2002/05/17 08:18:30 jrvosse Exp $