The Object Detector module implements functionality to load DriveNet and run several processing steps to detect all objects in the image. It is based on a deep neural network with image pre-processing and detection post-processing. The processing steps are illustrated in the following image.
For an example see Object Detector Workflow.