Artificial intelligence (AI) and facial recognition technology have been hailed as the future of crime-fighting worldwide. However, British police are emphasizing the synergy of these advanced tools with more traditional methods of detection. Which is the remarkable skills of super-recognizers.
Only one percent of the population possesses this extraordinary ability to remember and identify individuals with above-average accuracy. This is according to Tina Wallace, a surveillance expert with Thames Valley Police. In an effort to harness this unique talent, her team began recruiting specialist officers in 2017. It now boasts around 20 super-recognizers on their roster.
A super-recognizers online test that aims at soliciting the world’s best at facial recognition. Source| Scientific American
Among these exceptional officers is Alex Thorburn, a 17-year veteran of the force. Thorburn’s innate skill for recognizing faces prompted her to participate in tests conducted by the police.
“I’ve always been good with faces. So when they put a notice out about the tests, I did it!”
Alex Thorburn
In one of the tests, she was shown pictures of individuals, dated between 10 and 30 years old, and challenged to identify them in a crowded shopping center.
“I found them all, but they looked a lot different to how they did in the photos. That was really interesting.”
Alex Thorburn
The super-recognizers work with security camera footage on screens but are also deployed in the field for various assignments. For instance, during the coronation of King Charles III, Thorburn and her team were dispatched to Windsor Castle. They were tasked to assess the crowds and ensure the safety of the royal family.
Mike Neville, considers their role a “cheap and effective way to tackle crime.” Now retired, Neville, the founder of the first team of super-recognizers at London’s Metropolitan Police, runs Super Recognizers International. The organization claims to be “the world leaders in human recognition.”
One of the most notable successes of super-recognizers occurred during the 2011 London riots, which followed the death of a black man shot by police. Police had to sift through approximately 200,000 hours of security camera footage.
“Twenty officers identified 600 of the London rioters,” said Josh Davis, a professor in applied psychology at the University of Greenwich. Remarkably, one police officer who specialized in gang recognition recognized a staggering 180 offenders. The officer did so by analyzing the images, some of whom he had never seen in person.
Thames Valley Police, covering a vast area with a population of 2.34 million, has deployed its super-recognizers. They were strategically stationed outside bars and nightclubs to identify known sexual assault perpetrators.
Despite the rapid development of AI and facial recognition technology, human skills remain invaluable.
“It isn’t really a competition with facial recognition. They can be used together.”
Mike Neville
AI excels with high-quality, front-facing images, while humans excel with lower-quality images, angles, or partial face coverings.
Moreover, under UK and EU law, AI identifications must be verified by a human before an arrest is made. Neville noted,
“Most people, in a democracy, would be happier if people made the decision to detain someone, rather than a computer.”
Mike Neville
Demand for super-recognizers is growing, with police forces in Germany and Australia expressing interest. Professor Davis has even posted a 14-point test online for those curious about their own super-recognizer potential. Davis noted, “If you reach 14, contact me please!”
