This attribute reflects all the visual objects that have been recognized by the recognize operator, and thus have had their features integrated.
Substructure of the attribute:
(<top-state> ^visual-objects <vo>)
(<vo> ^feature <f1> <f2> ... <fn>
^loc-x <num1>
^loc-y <num2>
^loc-z <num3>)
^feature is a multiattribute listing all of the features of
the object that were integrated. The values are identifiers taken
from ^auditory-input,
^visual-input, or
^visual-input-peripheral.
The location attributes (which may be numbers in object-centered
coordinates, or the constant unknown) are used to determine
which features go together; no two visual-objects will share an
identical set of coordinates.