They only support 2D image recognition.

They can recognize and track complex 3D objects.

Model Targets are limited to specific predefined shapes.

Model Targets do not require any configuration.