The Video Labeler and Ground Truth Labeler (requires Automated Driving Toolbox™) apps enable you to create and import a custom automation algorithm to automatically label your data. Automation algorithms can be time-independent or time-dependent. Time-independent (nontemporal) algorithms can operate independently on each time stamp (or image). For example, a detection algorithm, such as the built-in People Detector, is a time-independent algorithm. In time-dependent (temporal) algorithms, there is a dependence on the time stamp of execution. For example, a tracking algorithm, such as the temporal built-in Point Tracker, uses tracking from a previous time stamp to track objects in the current time stamp.
If your algorithm is time-based, you must inherit from the vision.labeler.AutomationAlgorithm
and vision.labeler.mixin.Temporal
classes. For
example:
classdef MyCustomTemporalAlg < vision.labeler.AutomationAlgorithm && vision.labeler.mixin.Temporal
If your algorithm is time-independent, you only need to inherit from the vision.labeler.AutomationAlgorithm
class. For
example:
classdef MyCustomNonTemporalAlg < vision.labeler.AutomationAlgorithm
Inheriting from the temporal mixin class enables you to access properties such as
StartTime
, CurrentTime
and
EndTime
to design time-based algorithms. See the vision.labeler.mixin.Temporal
interface for details.
Only the Video Labeler and Ground Truth Labeler apps support both temporal and nontemporal automation algorithms. The Image Labeler app only supports nontemporal automation algorithms.
To create a temporal automation algorithm to use with the Ground Truth Labeler, open the app by typing
groundTruthLabeler
at the MATLAB® command prompt. Click Select Algorithm >
Add Algorithm > Create new
algorithm to open the template.
groundTruth
| groundTruthDataSource
| vision.labeler.AutomationAlgorithm
| vision.labeler.mixin.Temporal