Is facial recognition payment really safe?

Nowadays, besides scanning a QR code, the most used method of payment is facial recognition, but is it safe?

"Facial recognition" payment system is a payment platform based on facial recognition system, which does not require a wallet, credit card, or mobile phone. During the payment, you only need to face the camera on the POS machine screen, and the system will automatically associate the consumer's facial information with the individual account, making the entire transaction process very convenient.

With the maturing of facial recognition technology, "facial recognition payment" has gradually entered various aspects of our lives.

Although facial recognition payment is quick and convenient, it quickly makes people feel uneasy, always feeling there are security risks or other bugs. So, is facial recognition payment really safe?

The facial recognition technology we use on Alipay and WeChat Pay is 3D facial recognition technology, which is a bit safer than 2D facial recognition technology, especially in effectively avoiding impersonation situations brought about by various face forgeries.

Its principle is to perform liveness detection through a combination of software and hardware, in order to determine whether the collected face is a photo, video, or generated by software.

Moreover, when the 3D sensing camera recognizes a face, the built-in dot matrix projector can project more than 30,000 infrared dots invisible to the naked eye onto the user's face, thus providing richer data in terms of color, texture, depth, etc., making it safer.

In fact, what we are most worried about is that someone could use our photo to identify and steal our information and property, but actually

just using a photo is not enough.

Because there are three modes of liveness detection, the first type is silent liveness detection, to judge whether it's a real person, a photo, or a video.

The second type is recognition using visible light and near-infrared binocular cameras. Human faces have temperature. Under near-infrared light from 850 to 940 nanometers, the difference between human faces and photos/videos is significant. Through the two-dimensional image of human temperature and skin performance, the machine can make a judgment.

The third type is 3D face anti-forgery, through near-infrared plus three-dimensional structure recognition.

So, using a photo to impersonate is basically impossible, but this does not mean that facial recognition is completely invulnerable.

But to steal someone's information, to understand his account information, body information, and then launch a model attack, the cost is very high, and ordinary people can't do it.

Therefore, ordinary people don't need to worry about the safety of facial recognition technology.