Slashdot Mirror
Your Face or Fingerprint Could Soon Replace Your Plane Ticket (washingtonpost.com)
Headed on a trip? You may soon be able to ditch your boarding pass in favor of your fingers or face. From a report: Delta announced, on Wednesday, a new biometric identification pilot program that will eventually let you use your fingerprints instead of a plane ticket (Editor's note: the link could be paywalled; alternative source). That followed a JetBlue announcement hours earlier that it is testing a program in Boston that will match pictures of customers' faces with the passport database maintained by U.S. Custom and Border Protections. Delta's program, which kicked off at Washington's Reagan National Airport, is in partnership with Clear, a company that already lets customers skip to the front of security lines without identification.
Depending on the resolution of the camera and the training of the software's neural nets, the facial recognition can actually do much better than a person. There's a lot of stuff our brains just don't focus on, and there's a lot of detail that's too small for our eyes to really notice from any sort of distance, but a good enough camera will capture.
Of course also entirely possible for someone who knows enough about any particular software and how its been trained to trick it, as you can find plenty of demonstrations of if you google around a bit. And its also entirely possible for two people to really look similar enough that the software can't tell. And finally there will always be those edge cases that the software just wasn't trained on well enough and will confuse it.
But FR isn't really a joke anymore. We tend to think that people are 100% accurate at recognizing faces (police lineups and the such are based on this assumption) but the data shows that its really not true -- we're actually fairly bad at it.
But we have a few tricks to compensate: First, we only really pay attention to the faces of people we know. If we look at a picture with 10 people in it, we see John and Sarah whom we know, and 8 "others" that we don't know and don't care about. That's perfectly fine for our day to day activities where we mostly tend to ignore anyone we're not directly interacting with, but it doesn't do so great when you need to match a face against "one of 100,000 people."
The other trick we use is extra contextual information. If we know for sure that Sarah's at home and we go to the mall, any time we see someone who looks at Sarah, we can immediately shut it out because we know its not her (and of course if she changed her mind and DID come to the mall, this can lead us to not immediately recognizing someone we should be able to.) We can also use context such as knowing what kind of wardrobe our friends wear, the hairstyles they tend to prefer, etc. Our brains put all of this together to come up with a whole picture that just a headshot doesn't give us.
And of course if all else fails, we're really good at convincing ourselves its not our fault -- the light hit the person in a strange way or oh my goodness John has that exact same t-shirt or any of a dozen other excuses for why we just flat out got it wrong (we tend to do this for all failures of mental process of course, not just facial recognition -- our brains are wired to not admit our own faults.)
Anyway that turned out a lot more long-winded than I'd planned..