Lab 19b: Face detection

Kuri's default behavior is to roam around and snap photos, so you can imagine that Mayfield invested quite a bit of thought and effort into implementing some nice detectors. Amongst these out of the box detectors is a face detector which will likely come in handy when you implement interactions.

Note that these models are only available on the real robot. You won't be able to use the simulator to play with them.

Read Mayfield's Vision documentation

This is one area where the Kuri documentation does a pretty good job explaining things, so check out their page on vision for the outline of how the system works.

Mayfield vision documentation page

Test out face detection

• Create a script in applications called face_detection_demo.py. We've included all of the necessary message types in the course repo, so you should be able to run this node from the lab machine pointed at a robot master.
• Use the helpful Vision class in robot_api to activate the face_detector
• Register a callback on the face_change event and change the light color to visualize some aspect of the detections (e.g. confidence, number of faces, ID, size, position...)
• Use this demo to get a feel for how well the detector works. See if you can determine how far away a face has to be before it is no longer detectable, or other configurations that make it fail.