2.3 Principles for Creating MPEG-7 Documents

---

2.3 Principles for Creating MPEG-7 Documents

The user intending to use MPEG-7 for describing the content of multimedia data must be aware of two things: the conceptual model derived from the MPEG-7 requirements and the model implementation that leads to the content descriptions and the format of the descriptors/description schemes in the MPEG-7 DDL. Exhibit 2.4 shows the relationship between the conceptual model, the DDL, and the description schemes and descriptors.

Exhibit 2.4: Conceptual model and the description definition language.

For instance, assume that a user would like to create an MPEG-7 description for a video segment (part of the video) in which several image representations of this segment are available (in the form of a mosaic view). The user decides to use the VideoSegment DS for describing the video segments and the Mosaic DS for describing the mosaic view. Exhibit 2.5 shows the conceptual model for this example in Unified Modeling Language notation (note that the complete model for describing the content of video and images is not shown here, but will be discussed later). The VideoSegment DS entity inherits from a general and abstract entity, called Segment DS. The VideoSegment entity contains means for aggregating different Mosaic DSs. The Mosaic DS Entity inherits from the StillRegion DS entity, which is used for describing the content of still images.

Exhibit 2.5: Conceputal model in MPEG-7 for describing the mosaic images of a video segment.

The MPEG-7 DDL available comprises the VideoSegmentType, which is the XML Schema representation for the VideoSegment DS. It contains an element (which may have multiple occurrences to represent the aggregation) of MosaicType that corresponds to the Mosaic DS.

---