Posibilities have opened up for the devices to have context aware capabilities. The microphone, camera and other sensors that are built into the device can send the information which can be used for generating new outputs on the page, speaker and other connected devices. What this means to the user, programmer and finally the human kind? (why so Serious???)
Everyone has a finger or ten. Can we train the mobilenet model to recognize the finger? What can be done after that is the fun part... Do you see that pinkish box, can your finger move it with your finger?
MobileNet model
X Location
Y Location