Main Page
From Semantic Multimedia Wiki
Contents |
Welcome to the Semantic Multimedia Wiki
This wiki currently serves as a documentation platform for the K-Space Annotation Tool (KAT) and the Multimedia Metadata Ontology (M3O). The purpose of this wiki is not only to provide the documentation, but also to exchange ideas about semantic multimedia, and the hosted tools.
K-Space Annotation Tool
The K-Space Annotation Tool (KAT) is a framework for semi-automatic, semantic annotation of multimedia content. It is based on the Core Ontology for Multimedia (COMM) and provides a plugin infrastructure for easy extensibility. It further provides a GUI framework for adding visualization and user-interaction. A set of default plugins for manual image annotation and retrieval are included.
The KAT is developed within the K-Space Network of Excellence. The development is lead by the research group Information Systems and Semantic Web (IsWeb) of the University of Koblenz. If you would like to get involved, start to file bugs, add feature requests, publish your own branch and get in contact with the community or the core developers. Active contributors will be added to the core team upon request.
M3O - The Multimedia Metadata Ontology
With the Multimedia Metadata Ontology, short M3O, we aim at extracting the common structural information of today's multimedia metadata formats. The M3O is a follow-up initiative of the Core Ontology for Multimedia (COMM, http://comm.semanticweb.org/). Like COMM, it bases on DOLCE+DnS Ultralight and ontology design patterns. The major difference is that M30 is not focussed on a single metadata format like COMM is on MPEG-7. The goal is to provide a beautiful core ontology that is capable of expressing all structural information of today's multimedia metadata formats including the abilities of the COMM. However, it is easier to understand and to apply to other metadata formats, such as EXIF or IPTC. M30 consists of a set of core ontology design patterns used to express different types of metadata, decomposition of media, and semantic annotations. Based on requirements coming from the KAT, we will provide some instantiations of these patterns for the most common metadata requried, e.g., EXIF.
- Homepage of the M3O - Multimedia Meta-Metadata Ontology
Model-driven Generation of APIs for Ontologies
Existing tool support for generating application programming interfaces (APIs) for ontologies lacks sophisticated support for mapping the logics-based concepts of the ontology to an appropriate object-oriented implementation of the API. Such a mapping has to overcome the fundamental difference between the semantics described in the ontology and the pragmatics, i.e., structure, functionalities and behavior implemented in the API. Often concepts from the ontology are mapped one-to-one to classes in the targeted programming language. Such a mapping only produces concept representations but not an API at the desired level of granularity expected in a programming framework. We present a Model-Driven Engineering (MDE) process to generate customized APIs for ontologies. This API generation is based on semantics as defined in the ontology but also by additional information about ontology use. Such information can be described by ontology patterns and by further structures defined in this paper. This API generation process leverages the semantics specified in the input ontology and benefits from the use of ontology design patterns. The generated API serves as persistence layer for the pattern-based ontology and allows to extend existing classes with semantic information.
