WorldDimensionsMetadata
The name of a metadata field, in the training database, that describes the actual size of an object class.
The metadata field must contain three numbers - the height, length, and width of the object (in that order, separated by spaces). You can use any units, but you do not specify the units and you must use the same units for all perspective features. For example, to specify that objects in the class "car" are 1.55m high, 4.3m long, and 1.8m wide, the metadata field would contain the value 1.55 4.3 1.8
.
TIP: To add the metadata field to classes in your recognizer, use the action AddObjectClassMetadata.
Media Server can use this information to reduce the number of false positives by checking that a recognized object is the correct size. Media Server takes into account the position of the object in the scene and its distance from the camera.
You can model objects as cuboids, by setting a positive value for all three dimensions. If you set one or two of the dimensions to zero, the object is modeled as a rectangle or line. If you do not add the metadata field to an object class, or set all dimensions to zero, the object class is not used for estimating perspective and objects in that class are not filtered by MaxPerspectiveError.
For more information about using perspective in media analysis, see Use Perspective in Media Analysis.
Type: | String |
Default: | |
Required: | No |
Configuration Section: | TaskName |
Example: | WorldDimensionsMetadata=dimensions(m)
|
See Also: |