Dr. Manjeet Rege, the director of the Center for Applied Artificial Intelligence at the University of St. Thomas School of Engineering, spoke with WCCO-TV about how baseball fans entering Target Field now have an option for a faster line that allows cameras to scan their face instead of having to show a ticket.
"You are doing a kind of pre-screening beforehand," said Rege, who noted that the AI technology similar to MLB's new Go Ahead Entry system for facial recognition is also used to unlock phones, at airport security check points, and in forthcoming capacities like remembering preference settings for a car's driver.
From the article:
"The accuracy of these recognition models have gone up significantly," said Dr. Manjeet Rege. "The one that is being used in Go Ahead lanes is using NEC's NeoFace technology that has an accuracy of 99.85%."
Since the tickets are attached only to the ticketholder — the customer using the Ballpark app — it can eliminate fraud from a stolen ticket.
Concerns that pictures could be stored or be incorrectly identified stand out. Studies have shown that facial recognition is less accurate for people of color.
But Rege feels recent improvements to AI algorithms have reduced issues of bias, including for Go Ahead entry.
