in

Intel brings FaceID-like biometric authentication system to ATMs


Intel on Thursday announced a new facial recognition system that promises to bring FaceID-like biometric access to ATMs, smart locks and more. The Santa Clara, California-based chipset giant said its new RealSense ID camera system combines an active depth sensor with a specialized neural network designed to deliver secure, accurate and user-aware facial authentication. The RealSense ID system starts at $99 and will be made available in Q1 2021.

The idea behind the RealSense ID camera system is simple. Due to the pandemic, whenever a consumer hit an ATM, there is a greater risk of coronavirus infection. But more than that, Intel believes that its RealSense 3D camera finally brings facial authentication on consumer-facing devices such as point-of-sale systems, ATMs, which have been neglected for years.

Similar to FaceID on Apple’s iPhones and iPads, RealSense ID takes advantage of two camera lenses and special sensors that can capture depth. Combined with a specialized neural network and a dedicated system-on-chip it can detect and distinguish a person’s face.

Apple FaceID tips and tricks you did not know about

Intel says the system can adapt to users’ changing appearance over time. The solution also works in various lighting conditions for people with a wide range of heights or complexions.

The company said its RealSense ID has built-in anti-spoofing technology that should block any attempt of false entry through the use of photographs, videos, or masks. RealSense ID has a one-in-one-million false acceptance rate.

“The solution is also only activated through user awareness and will not authenticate unless prompted by a pre-registered user,” Intel said. “As with all Intel technology, we are working to ensure the ethical application of RealSense and the protection of human rights.”

Intel has been facing tough competition from both Apple and AMD in the chip business. The company first introduced its RealSense 3D camera technology in 2014.





Source link

International team of researchers uses photonic networks for pattern recognition — ScienceDaily

A better pen-and-ink system for drawing flexible circuits — ScienceDaily