Join or Sign in

Register for your free membership or if you are already a member,
sign in using your preferred method below.

To check your latest product inquiries, manage newsletter preference, update personal / company profile, or download member-exclusive reports, log in to your account now!
Login asmag.comMember Registration

Unveiling the many faces of facial recognition

Unveiling the many faces of facial recognition attends a Solutions Day event by CyberLink, which displayed its FaceMe series of solutions and showed how they could be applied in security and non-security use cases. attends a Solutions Day event by CyberLink, which displayed its FaceMe series of solutions and showed how they could be applied in security and non-security use cases.
Facial recognition has seen increased adoption, and its growth is expected to continue in the near term. According to Allied Market Research, the global facial recognition market size was valued at US$3.83 billion in 2020 and is projected to reach $16.74 billion by 2030, growing at a compound annual growth rate of 16 percent from 2021 to 2030.
There are many reasons for this, one being facial recognition’s contactless nature, which attracts users in the midst of the pandemic. Another reason is growing confidence of the technology as an effective authentication and management tool. Prior concerns that facial recognition has bias against certain ethnic groups have diminished as the technology becomes more accurate.
“What we’ve seen over the last five years … during the first four years facial recognition was mostly talk, yet it witnessed significant growth during last year,” said Jau Huang, Chairman and CEO of CyberLink, during the event in Taipei.
“We have business in Taiwan, Japan, Korea and Southeast Asia, among others. US had been the slowest adopter. Yet facial recognition witnessed a dramatic increase in the U.S. recently,” he continued. “The U.S. had always expressed privacy and accuracy concerns, saying error rates among certain ethnic groups were higher than the others. But you can see that these concerns are no longer voiced so much by critics.”
The company cited their own example of having a 99.73 percent accuracy rate when a visa photo is matched against a database, and 97 percent accuracy rate with a “wild photo” – a poorer-quality image taken by IP cameras or other sources. The company’s facial recognition has been recognized by renowned organizations such as National Institute of Standards and Technology and iBeta.

Giving security a boost

Indeed, security is still a major application of facial recognition. In this regard, CyberLink has a solution, FaceMe Security, that leverages users’ existing equipment and hardware. “The user might have built their video surveillance system years ago; do they need to change their system to use FaceMe Securtiy? No. This is our core value, to allow users to use their existing system,” said Jimmy Luo, Product Manager at CyberLink.
Among FaceMe Security’s advantages cited by Luo are accuracy, as mentioned earlier, and scalability, allowing users to choose the best solution according to their business size and requirements, from 4 cameras that can be supported with an Intel NUC to over 60 cameras supported by NVIDIA’s RTX A5000 GPU along with a workstation.
Plus, FaceMe Security integrates with six VMSes including Genetec, Nx, Axis, Avigilon, Milestone and VIVOTEK, all of which covering a majority of the VMS market. Among them, Genetec is the VMS with which FaceMe Security has deeper integration; all data on Genetec can be synced instantly to FaceMe Security upon integration. FaceMe detection events can be imported directly to the Genetec interface. Smart search of all footage of an individual in question across different cameras can be achieved, sharply reducing search time.

Non-security applications

Besides security, facial recognition applications have been extended to non-security areas as well. CyberLink showed some real-life projects where FaceMe is used to facilitate users in this regard.
In retail, for example, FaceMe has been integrated with POS machines that authenticate shoppers as they check out goods. The shopper shows their face and the payment process is approved. FaceMe also includes anti-spoofing features where shoppers are denied payment if they try to cheat the system with a photo of someone else.
Another case is in a Toyota plant, where certain operations can be performed only by those with permission. In this regard, FaceMe is used to authenticate workers, who can be vetted without having to take off their mask, goggles or hard hat. “This way workers can concentrate on their work, and health and safety requirements are met,” said Mei Guu, Senior VP of CyberLink.
Finally, in “Fintech,” facial recognition again provides an effective means for authentication. FaceMe, for example, can help authenticate users seeking to remotely open an account with a financial service. The user can take a photo of their ID and then a selfie to be processed by the FaceMe app, which can then establish if there’s a match. Once approved the user can open an account, a process that has been reduced to a few minutes from days had the application been processed manually by person.

Product Adopted:
Subscribe to Newsletter
Stay updated with the latest trends and technologies in physical security

Share to: