The example can draw a box of the object and show the predicted label. It can interact with the WebRTC or the visual content uploaded by user.