Find all our latest news, insights, and events

Face ID Algorithm Grants Access With Smiles & Winks

Blog Image

Posted: Jul 26, 2021 | Reading Time: 2 Minutes


A new facial recognition algorithm has the ability to not only use a person’s face as an ID, but also gestures such as unique facial motions, or lip movements. This new form of biometric identification can make it even harder for hackers and thieves to use a person’s regular facial identity.

Faces in motion

Created by Professor D.J. Lee of Brigham Young University in Utah, the feature is called Concurrent Two-Factor Identity Verification (C2FIV), and requires both a user’s facial identity along with a specific facial motion to gain access to a device. To set up, the user must face the device’s camera while it records a one to two second video of their facial gesture of choice, whether it’s a wink, a smile, or a specific lip movement from saying a certain word or phrase. The device then analyzes the facial and motion features of the video, which is then stored for future identification. The main issue with regular facial identification is that hackers or thieves can simply use images taken of a person’s face from social media or even while they’re unconscious to access their devices.

Technical workings

C2FIV is dependent on an integrated neural network framework to learn both facial features and facial movements at the same time. Instead of focusing on a single image as regular facial recognition does, the framework takes every frame of the initial facial recording into consideration when reading a user’s facial movements. The user’s facial features and movements are either stored on a server or embedded into the device for when access is needed, and the device compares the saved data to the data that is input by the user to allow access. In preliminary studies, Lee and a Ph.D. student recorded 8,000 different video clips from 50 subjects that made facial movements such as blinking, smiling, or raising their eyebrows, along with many random facial motions to train the C2FIV algorithm. The algorithm was able to identify facial movements correctly with over 90% accuracy.

Lee has already filed a patent for the C2FIV technology, and hopes to be able to apply it to broad areas as opposed to phones and laptops, such as workplaces, online banking, ATM use, or even for room entry at hotels or vehicle access without a car key. "How great would it be to know that even if you lost your car key, no one can steal your vehicle because they don't know your secret facial action?"

ClinTech Systems specializes in healthcare and business informational technology. We are the leading experts in HIPPA, PHI software compliance, integration of CRM, EMR, HRIS, Data Relays, Office 365, managed information technology systems, and offer CIO consultation to increase efficiency and enhance productivity. Our staff is highly focused on leading strategic innovations, troubleshooting complex problems, and building bulletproof solutions.

Back to Blogs
shape image

Let's talk about your project and see how we can work together

We're a team of Engineers, Builders, Network Experts, and Creative Programmers. So, what can we create for you today?

Contact Us
icon
icon
icon
icon
icon
icon
icon
icon
icon