Covid-safe, real-time facial recognition is set to reduce queues at the Tokyo Olympics.
Athletes at the Tokyo Olympics will rely on computers to automatically match their faces to their photo to ensure they don’t show up too late at the start this summer.
Tournament security is enhanced by real-time facial recognition technology in venues where space is too limited for manual identity checks and the queues that these procedures often create.
More than 40 venues, including the main stadium, the International Broadcast Center and the Olympic Village, will be covered by the system, which is designed to ensure that more than 300,000 accredited athletes, staff, media and volunteers can be quickly verified and admitted .
It won’t completely replace physical checks and compares a photo ID embedded in a traditional lanyard security pass to the face of the person seeking access, most likely with greater speed and accuracy than human personnel could achieve for the same comparison.
Rather than focusing on a single, centralized Olympic park, Tokyo’s sports venues are scattered throughout the city. Athletes must be able to travel freely between these locations and authenticate at each location, and the potential for delays that can affect a tight schedule of events is significant.
Organizers are also concerned about the health consequences of people queuing for too long in the summer heat (the average July temperature in Tokyo is between 23 and 30°C).
Facial recognition has another distinct advantage over other forms of biometric identification (such as fingerprint recognition) in that it is contactless, doubly important to the Tokyo 2020 organizing committee this year as it reduces the chances of coronavirus infection caused by touching shared surfaces.
The system was previously piloted at the 2016 Games in Rio de Janeiro and is based on NeoFace Watch technology developed by Japanese IT and electronics giant NEC.
NeoFace Watch uses the generalized face detection method, which uses an artificial neural network model, the generalized learning vector quantization algorithm, that searches and selects facial features for attunement purposes.
The perturbation space method converts two-dimensional images such as photographs to three dimensions — a process called morphing — so that the candidate’s head can be turned left and right and up and down to achieve greater accuracy.
Further processing applies different levels of lighting filters over the subject’s face to facilitate an even better match of the photo with an image stored in a central database.
‘All personal data will be properly managed and used during the competitions and then securely deleted under strict conditions.
‘ Official statement from the organizers of the Games
Accurate facial recognition can still be hampered by changing expressions – such as a smile or accidental blinking – and wearing head accessories such as hats or sunglasses. To counteract those effects, NeoFace Watch uses an additional process called the adaptive regional blend-matching method to minimize the impact of local changes.
NeoFace Watch is highly rated by the US National Institute of Standards and Technology (NIST), which measured its ability to match facial image data against large databases of people in personal identity surveys, as well as ‘one-to-one’ matching capabilities where two facial images are used.
evaluated to determine if they are the same person. Subsequent NIST testing suggests the system can check 1.6 million faces in just 0.3 seconds, with an estimated accuracy of 99.7 percent. It’s not clear whether the processing will be that fast in Tokyo, or whether it will take place on the hundreds of high-definition video cameras installed around Tokyo sites that monitor access facilities.
The tampering is more likely to occur on attached peripherals, such as web-based thin client terminals powered by Intel processors. The chipmaker launched its own RealSense ID system in January of this year.
This uses a small module with a special system-on-a-chip component to securely process and encrypt user data, which can be easily integrated into other products, supplied as a stand-alone peripheral that can be run on any computer.
plugged in. Concerns about data privacy have inevitably grown, not least because NeoFace Watch is just one part of NEC’s Bio-Idiom line of biometric authentication technology, which has been widely embraced by criminal surveillance and other enforcement organizations around the world.
For example, it’s currently used at New York’s JFK Airport, where it’s used to match faces against a photo database of “junk” items and to send a notification when a match is found.
Another deployment is with the New South Wales Police Department in Australia, which uses the system to match real-time CCTV images from live camera feeds to a watchlist of individuals of interest, and to compare crime scene images with those stored in databases to identify matches.
The Merit Lefkosa Casino in Northern Cyprus uses NeoFace Watch to track the activities of ‘problem’ gamblers and unauthorized individuals who are not already whitelisted, while the same system identifies VIPs to give them faster access to the gaming rooms.
Reports suggesting the system will also be used to provide contact tracing capabilities to help monitor the spread of coronavirus during the Games remain unconfirmed, although we know that NEC has already developed the technology to monitor people and groups for compliance with guidelines for transmission reduction.
For example, NeoFace Watch Thermal combines facial recognition and thermal imaging technology to make locations Covid-19 safer using specially equipped CCTV cameras, which also record the date and time of a person’s visit to a specific location.
The Tokyo 2020 organization has not said how long it will keep people’s personal data, but has promised to comply with local privacy laws.
In Japan, this falls under the auspices of the Personal Information Protection Act, originally passed in 2003 and last amended in 2020 (the government wisely mandated in 2015 to consider updates every three years to keep up with the rapid evolution of technology and global standards).
“All personal data will be properly managed and used during the competitions and then securely deleted under strict conditions,” the organizers say.
NeoFace Watch will only be applied to athletes, staff, media and volunteers – who have given pre-approval when submitting their photos for inclusion in the central database and on the ID card alongside a built-in IC chip with their personal ID and authentication information.
The organizers insist that it will not be used for the general public or spectators who have not given the same express permission.
In the unlikely event that the data is not collected, stored and processed correctly and legally, in accordance with Japanese privacy law, it may take some time to be released.
But what may be harder to hide under the intense media attention are queues outside of venues if the accuracy of the NeoFace Watch system falls short of organizers’ expectations.
On very public test is not only the technical prowess of the athletes themselves, but also the efficiency of AI-powered facial recognition systems used for ID checks and queue management, which could lead to extensive implementations in other scenarios long after the closing ceremony is completed.