Device-to-device audio communication (think QR codes, but made out of sound) seems very underrated. Much better UX than QR codes or NFC because you don't have to worry about putting your phone in the exactly right place or pointing it in the right direction.

Anyone here with more detailed takes on this?

12 years ago I worked on a using multiple sensors called "Proxignos" — I know horrible name. The idea was to use multiple mobile sensors to statistically improve probability that two devices were in the same location using GPS, accelerometer, camera, a 5 second audio clip, and device time.