Estimates show that the facial recognition market will hit a stunning $12 billion market cap by the year 2025. Thus, the industry is expected to grow, find new use cases, and come down to more commercial users in the space of five years.
If that is true, this is the time to have a much-needed discussion about this relative newcomer to the biometrics market. We do hope that you are ready for this discussion too.
What are Facial Recognition Systems?
It is easy to see this as a system that takes a face and notes that it is yours or not. What goes on behind the scenes is more complicated than that. This piece is not a science paper, though, so we will try to break it down until it gets as simple as it can be.
As with any other system that is made to compare anything, a sample has to be provided. This could be a primary sample that is fed into the system by a human or automatically captured image by cameras.
No matter how the facial data is obtained, it is usually in analog form when it comes in. To understand this face better, the system observes unique features for a digital print.
Here, the facial recognition system takes notice of unique features such as:
- The distance between the lips
- The distance between the forehead and the nose
- The distance between the forehead and the lips
- The general shape of the face, etc.
All of this data forms a digital signature for the face, which is then stored in the system.
If that is the first time when such a face will be run through a system, there would be no matches. After all, it will just come in as a fresh sample. Should the same face be run into the system for a second time, though, it is bound to return results matching the digital print of such a face.
With this model of operation, the facial recognition system is built to ask and answer two different questions. They are:
- Who are you? Here, the system aims to identify the user/ person who’s facial data is being run through it.
- Are you who you say you are? This is the authentication phase, where the system still has to confirm if your digital signature matches what it has stored.
Why Facial Recognition Systems are all the Fad
Before considering any other thing, it is almost very natural that facial recognition systems get the kind of acclaim which they have today.
Of course, we have other forms of biometrics setups. This has given us palm vein digitization, voice models, iris scans, fingerprint systems, and much more. However, we recognize each other with our faces – so it is natural that a system is built to do the same thing.
In line with that, facial recognition systems are as contactless as they can be.
Powerful cameras that can capture and identify faces to a high level of accuracy, even from a very far distance, are deployed everywhere today. So, the person does not also need to come close to the camera before they could get identified.
Looking around us today, facial recognition is a practical tool to prevent viral infection and pandemic. After all, such could quickly spread by systems requiring contact.
Besides that, we cannot deny the fact that these systems work.
There was a time when the facial systems could only capture the face in 2D. This worked, but there were a lot of flaws in it. Where a simple image/ picture the person could be used to fool such systems, it became less desirable.
Fast forward to today when there are 3D systems in commercial quantities, and we have a more robust security model.
So available has the top-notch facial recognition system been that we now see it on the iPhone X series of devices – and every other phone that the Cupertino-company has built in the same form factor. Although they came under fire when the earlier models faulted, they have since improved their offering with time.
Security is also at the heart of facial recognition.
Arrests have been made based on this system alone in developed parts of the world. Although inconclusive, studies have shown that communities and neighborhoods with similar systems involved tend to see lesser crime.
A more worthy pick is that of 3,000 children all over India that were found and reunited with their families in less than four days.
The Grey Side
There is rarely one piece of technology that wins all round, and facial recognition systems are not left out of that mix.
Till today, the biggest concern with deploying such systems in public is consent.
When we have this piece of technology on our phones, we have the option to choose to use it or not. Thus, the capture and use of our facial data are within our control. When in public, though, who consents to your facial data being captured and used by the law enforcement agents, businesses, etc.?
It is even harder because we cannot leave our faces at home. While we can use methods like using a VPN, keeping photos offline, and using pseudonyms to stay private, none of these will work for facial scanning systems. A more serious approach would be to subscribe to the use of privacy goggles that keeps your face hidden from these cameras.
But then, for how long do we keep doing that?
Even if we did not have a problem with facial data being collected, we would want to know how this data is being managed. What happens when a leak occurs, and your digital face signature is out there in the wild?
Now that it is sure that we will see more of these system models in the future, can’t someone impersonate you with this facial print?
This is why legislations have started springing up all over the world to either ban/ regulate the usage of this technology. Until a balance can be struck between its strong points and concerns for user privacy, though, the use of this form of technology in public spaces might not gain the full acceptance it seeks.