Stephen Hyduchak
Jul 28 · 4 min read
Vist www.goaver.com for more information and blog

Buzzwords are often on minds during “hot” times. Things like blockchain, AI, and facial recognition are some to name a few; but for this intent we are going to dig into the differences between facial recognition and authentication.

Let’s break it down from a high level:

Facial authentication: This is considered a form of biometric authentication by definition and looks at unique biological characteristics so identity can be attested.

Facial Recognition: This is a software driven application that looks at facial points and contours to analyze and compare to make the attestation of like-ness.

Which are used and where?

They can be used either separately or together and it is best to look at industries that each are applied too.

Facial Recognition

1) A good example is Facebook, we all have seen the box come up around faces or friends and family in our pictures and the “auto-tag” of that user.

2) Casinos: “The eye in the sky” as they say when you are in Vegas. Most resorts have this in place to scan faces as they enter the casino. This is done to stop people on voluntary exclusion lists or known cheaters that were blacklisted.

3) Financial Services (banks, cryptocurrency exchanges): Facial recognition is used for account openings and online account access.

4) Retail: Face recognition is applied to know when shoplifters or retail criminals enter establishments with history of theft.

Next lets look at the process flow and how facial recognition works:

1) The image is captured in the form of a selfie or from a crowd, the more direct on then the better the results for analysis.

2) Microsoft Cognitive Services say they use ~30 facial points for each face in the image, as well as attributes that include age, gender, emotion, pose, smile and facial hair. This is also known as the “T” zone of the human face.

3) Next the image is compared to the database it is being applied too; this can be a document like in know-your-customer (kyc) processes.

4) Decision time: All the variables that were mentioned in step 2 can be some in the codebase, but what comes out is the percentage to match images. The systems are evolving and becoming smarter, but by no means foolproof. Lighting, distance and camera quality are just a few things to really throw false positives.

Facial Authentication

The easiest way to think of this is when you unlock your Apple Iphone or Android Device.

You just received that shiny new device (hopefully purchased with cryptocurrency) and are prompted through the oblong-shaped circle to put your head onto the screen, while the facial points are made and stored. After these points are recorded centrally on that device; every time you unlock your phone with your face, this becomes your password or authentication. Makes sense?… We thought that was easier to picture than the technical jargon!

Now we are getting somewhere, facial recognition drives the user facial authentication in a few steps. Lets look at it from a process we have all most likely done before (KYC):

1) Capture a picture from driver license or identification card

2) Take a selfie, using technology that ensures it is not stored and live at the moment helps spoofing and false positives

3) The facial recognition goes to work and compares the images through algorithms and renderings

4) Confirm the digital identity and the customer is allowed access.

The SMS Fail

No, this is not a Fortnite dance. Many businesses over the last decade have relied on two-factor-authentication (2FA). This is when you receive a code on your mobile device to continue access. Due to “sim swapping” and phishing attacks, this is no longer considered best practices.

The Good News

Thanks to Apple and other big technology hardware providers, Face ID has paved the way for trust and confidence in this technology. Ease-of-use drives consumer and customer behavior and the numbers do not lie, this is the preferred way rather than one-hundred passwords in your notebook.


For businesses, especially in emerging markets like cryptocurrency and gaming that have younger user bases, this is the technology customers demand. They want a streamlined process, with their consent and to gain access into their favorite platform or trading exchange. Aver is using the latest technology on the back of Microsoft Azure, so you can trust your facial systems are secure, web-based and the most-feature rich on the market. It is important to understand the differences and learn how these can meet the demands of evolving businesses.

Aver

Digital Identification Through AI

Thanks to Bridge Protocol

Stephen Hyduchak

Written by

Blockchain, Identity Verification, AI and Cannabis keep me up at night. CEO of Bridge Protocol and Aver.

Aver

Aver

Digital Identification Through AI

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade