Publication,
Research,
Docs and Info,
KAMEDA
A dynamic situation is the key concept for the video imaging. It indicates a status of the real space at a time and is defined by the strategy of what kind of video is desired and how it is generated. Hence it depends on the activities or the contents.
The dynamic situation is represented in two ways; the symbolic representation for the user and the feature representation for the system. The advantage of this representation is to avoid the pattern recognition problem. In other words, a user takes advantage of the symbolic representation, while the system takes advantage of the feature representation. Hence the description has two layers. The bridge between the two layers is the two way representation of the dynamic situation.
In the symbolic representation, the dynamic situation is described by a set of actions of the active objects. A description of this action is called action component and is denoted by A-component for short.
On the other hand, in the feature representation, the dynamic situation is defined by the combination of the features extracted from the sensor data. We call these features the situation features.