Chinese ‘gait recognition’ tech IDs people by how they walk
BEIJING — Chinese authorities have begun deploying a new surveillance tool: “gait recognition” software that uses people’s body shapes and how they walk to identify them, even when their faces are hidden from cameras.
Already used by police on the streets of Beijing and Shanghai, “gait recognition” is part of a push across China to develop artificial-intelligence and datadriven surveillance that is raising concern about how far the technology will go.
Huang Yongzhen, the CEO of Watrix, said that its system can identify people from up to 50 meters (165 feet) away, even with their back turned or face covered. This can fill a gap in facial recognition, which needs closeup, high-resolution images of a person’s face to work.
“You don’t need people’s cooperation for us to be able to recognize their identity,” Huang said in an interview in his Beijing office. “Gait analysis can’t be fooled by simply limping, walking with splayed feet or hunching over, because we’re analyzing all the features of an entire body.”
Watrix announced last month that it had raised 100 million yuan ($14.5 million) to accelerate the development and sale of its gait recognition technology, according to Chinese media reports.
Chinese police are using facial recognition to identify people in crowds and nab jaywalkers, and are developing an integrated national system of surveillance camera data. Not everyone is comfortable with gait recognition’s use.
Security officials in China’s far-western province of Xinjiang, a region whose Muslim population is already subject to intense surveillance and control, have expressed interest in the software.
Shi Shusi, a Chinese columnist and commentator, says it’s unsurprising that the technology is catching on in China faster than the rest of the world because of Beijing’s emphasis on social control.
“Using biometric recognition to maintain social stability and manage society is an unstoppable trend,” he said. “It’s great business.”
The technology isn’t new. Scientists in Japan, the United Kingdom and the U.S. Defense Information Systems Agency have been researching gait recognition for over a decade, trying different ways to overcome skepticism that people could be recognized by the way they walk. Professors from Osaka University have worked with Japan’s National Police Agency to use gait recognition software on a pilot basis since 2013.
But few have tried to commercialize gait recognition. Israelbased FST Biometrics shut down earlier this year amid company infighting after encountering technical difficulties with its products, according to former advisory board member Gabriel Tal.
“It’s more complex than other biometrics, computationally,” said Mark Nixon, a leading expert on gait recognition at the University of Southampton in Britain. “It takes bigger computers to do gait because you need a sequence of images rather than a single image.”
Watrix’s software extracts a person’s silhouette from video and analyzes the silhouette’s movement to create a model of the way the person walks. It isn’t capable of identifying people in real-time yet. Users must upload video into the program, which takes about 10 minutes to search through an hour of video. It doesn’t require special cameras — the software can use footage from surveillance cameras to analyze gait.
Huang, a former researcher, said he left academia to co-found Watrix in 2016 after seeing how promising the technology had become. The company was incubated by the Chinese Academy of Sciences. Though the software isn’t as good as facial recognition, Huang said its 94 percent accuracy rate is good enough for commercial use.
He envisions gait recognition being used alongside face-scanning software.
Beyond surveillance, Huang says gait recognition can also be used to spot people in distress such as elderly individuals who have fallen down. Nixon believes that the technology can make life safer and more convenient.
“People still don’t recognize they can be recognized by their gait, whereas everybody knows you can be recognized by your face,” Nixon said. “We believe you are totally unique in the way you walk.”
local mobile number and a Chinese bank account.
One afternoon they were desperate for caffeine and spotted a Luckin Coffee outlet, an up-andcoming Chinese brand, in the canteen of Bytedance, the AI information and entertainment powerhouse. But Luckin takes orders only on its mobile app. At the cashier-less convenience store at the headquarters of JD.com, the online retailer, an employee paid for their snacks with his own phone.
“China’s internet is a walled garden,” Chan said. “No one can break in unless you’re from here.”
Within that walled garden, everything seemed to be moving at an extraordinary speed. While Silicon Valley startups raise funding every 18 to 24 months on average, the group was told that the most successful Chinese companies do it every six months. It isn’t unusual for a hot startup to raise funding three to four times a year.
“Every time I go to the U.S., I feel that I’ll need to grow 10 times faster,” said Alexander Weidauer, a founder of the Berlin-based AI chatbot developer Rasa and the only member of the group not from Silicon Valley. “Now I feel I’ll need to grow 100 times faster. The pace in China is crazy.”
Their hosts kept reminding them of the advantages China had over the United States in AI development. China’s vast population and loose privacy laws give them access to much more data. AI companies also have considerable government support and are willing to pay more for top talent.
“The U.S. competitive edge over China may not be long,” Kai-Fu Lee, chief executive of Sinovation Ventures and former head of Google China, told them. “In fact,” he added, “the Americans now have the information disadvantage.”
But it was also obvious to the group what China was missing. For starters, everybody is Chinese. Even in its early days, Google had employees from 39 nationalities speaking 40-plus languages.
“China is a bit homogeneous,” said Chan, an early Google employee. “You don’t find as much the perspectives of the world here compared to Silicon Valley.”
Then there are the work schedules. The Silicon Valley natives were introduced to the Chinese startup concept of 996: Work from 9 a.m. to 9 p.m., six days a week. Once they got over their shock, they had to ask: Does that punishing schedule make sense?
“I’m not worried so much about my portfolio companies not working as hard as the Chinese companies,” said Chan, now a partner at Felicis Ventures. “I’ll worry when they’re less creative and less efficient.”
While China is becoming more innovative, many members of the group said they believed — and some of their Chinese counterparts agreed — that the United States still led in some areas.
Although some Chinese tech companies can look very Silicon Valley-esque — with sprawling campuses that include dining halls, gyms and nap rooms — their preferred management style is still top down and results driven. Unlike Silicon Valley, smart underlings have less freedom to start something new.
Underscoring their admiration for strong leaders, the Chinese technology figures told the Silicon Valley group that many in China idolized Uber founder Travis Kalanick, who resigned as chief executive last year after the company was embroiled in various scandals. The reason: In his battles with a bigger local rival, Kalanick could be as aggressive and scrappy as a Chinese boss.
There was very little discussion about the consequences of Chinese companies’ ruthless focus on growth and the social impact of the technologies they develop — criticism Silicon Valley now faces. The visitors asked how Chinese companies dealt with the issues of censorship and algorithm-driven social media, but their hosts either seemed puzzled by the questions or brushed them off.
Technology itself is neutral, some of the Chinese executives said. It depends on how people use it — an argument that Silicon Valley companies used to make.
And then there was the surveillance. Chinese companies have little choice but to cooperate with Beijing’s growing efforts to track the daily lives of its own people. Some, in fact, make money off it.
Still, the Silicon Valley delegation did not expect facial recognition technology to be so widespread in Chinese life. Companies they visited used it at office entrances and at retail kiosks inside their facilities. They also saw demo videos of how the Chinese police could use the technology to monitor potential crimes in crowded public spaces and learn how many suspects had been arrested.
After grasping how prevalent the cameras were, they started counting them. Even the van they rented from the hotel in Shenzhen had a handful of cameras installed, blinking from time to time to signal that they were on.
Then, like many people in China, they got used to it, as if the cameras weren’t even there.