AI Inspiration #10: Moneyball 3.0; Fishy Facial Recognition at Airport Security; Image Recognition for Adult Videos
Here’s everything that’s new in artificial intelligence and computer vision, with a little tech pop culture to make the medicine go down. Our logic is undeniable.
But first, let me take a selfie.
Just in case nobody ever told you, touching or talking to the animals that appear in your selfies isn’t good for the animals. Suspecting a disturbing trend in the practice, two charities turned to image recognition and machine learning to pick out “good” from “bad” wildlife selfies on social media. Their findings? There’s been a 292 percent increase in wildlife selfies on Instagram since 2014, and 40 percent of those are considered of the “bad” variety. The hope is that a little bit of said machine learning could be used to train some humans in better tourist tendencies.
Security areas at Dubai International Airport are about to become fun thanks to computer vision. The airport is installing virtual aquarium tunnels featuring video footage of fish swimming all around travelers. The entertainment will draw travelers’ eyes all over the screen, enabling the 80 cameras embedded in those screens to properly scan their faces, and, eventually, look for explosives or other dangerous materials.
From increased HD cameras across arenas to drones to in-ball sensors, the proliferation of visual data capture methods at sports events is combining with advances in AI to create new ways for teams to optimize their gameplay strategies. Welcome to Moneyball 3.0.
NSFW adult entertainment aggregator Pornhub is applying computer vision to help tag everything from actors to specific sex acts in its pro and amateur videos. The site gets more than 10,000 clips uploaded by users every day, and this new initiative will certainly make it easier for viewers to find what they are looking for, whether the people in the videos want to be found or not.
Addresses and “we missed you” delivery slips will soon be a thing of the past if the DelivAir concept ever sees the light of day. This proposed drone delivery service uses your phone to find out where in the world you exactly are, then delivers your package directly to you after it verifies your identity via facial recognition and a QR code sent to your smartphone.
Anyone you know interested in computer vision? Forward this to them so they can subscribe, too. And please submit any computer vision stories you think we’d be interested in posting.
The Visionary newsletter is produced by GumGum.