They remind me of a uni project a guy was doing in 2010. Machine and human readable cubes to allow a machine vision and computer to read communication lines from a human.

You'd place the cubes in a row to spell out a sentence of instruction. And the camera capture would send to be deciphered.

At the time I couldn't comprehend how this was useful, but maybe specific graphics were easier for the machine vision to recognize then actual words.

He tried explaining it to me, but after a few times I said something like "I'm either dumb as a brick and can't get this or you have gone off the derpend."