Description
While looking into the world eco system I am trying to find how the iris print is verified. I was hoping for maybe some ~40 bit unique number which could then be used to feed into a recovery algrithm as the whitepaper seemed to suggest that there are some 2^46 equally unique iris picture comparison results maybe for each eye, maybe some a slightly more common, so lets say 40bit. Even if the whitepaper states:
Currently, the error rate of the Orb for confusing any two people to be the same, is approximately 1 in 40 trillion. On a billion people scale, this translates to a [99.999% true acceptance rate](https://world.org/blog/engineering/biometric-performance-billion-person-scale) or 0.001% false rejection rate, which is significantly better than other known alternatives.
But so far I only have found this fuzzy matching algorithm of two pictures. You can't be serious right? The whole premise of the Orb placed on interference, so by definition an inexact science? Or is this just an example library for third party implementations which need a fast open source image matching algrorithm that is trained on iris pictures? It would be great if the purpose of this repo / code within the worldcoin ecosystem could be clarified in the README.
Thanks!