Top 9 Facial Recognition Technology Trends of 2022

The facial recognition market is expected to increase by $9.6 billion by 2022. This will be driven by its applicability to an ever-widening variety of solutions. Today, face recognition technology is preferred over other types of biometric technologies such as voice fingerprint scanning, voice recognition, texture recognition and skin texture recognition due to its contactless nature and straightforward deployment. Currently, facial recognition is predominantly used in security and marketing.

In this article, we’ll zoom in on the most popular facial recognition trends in 2022 and find out the most interesting cases where this technology can be used. But before we start, let’s quickly look at the current status of facial recognition technology and its market position.

Contents

• What is facial recognition technology, and how does it work?
• Global facial recognition market
• Uses of facial recognition technology
• Facial recognition trends
• The future of facial recognition

What is facial recognition technology, and how does it work?

Face recognition is one of the ways to identify and confirm a person’s identity in photos and videos, including in real time. Like fingerprint and voice recognition, it is a form of biometric technology. Face recognition consists of:

• Face detection At this stage, the camera captures a face from a photo or video. Then the camera draws a box around the face to autofocus it. The main purpose of this phase is determining the presence and location of a face.
• Face analysis: The face recognition technology maps faces and measures the distance between the most important face landmarks. These calculations are later transmitted into a string of numbers known as a “faceprint” or “face signature.” The faceprint is added to a recognition database.
• Face recognition: The face signature is compared to a database of known faces. This process helps to confirm the identity of a person in a photo or video. The richer the diversity of photos embedded into the system, the more accurate results we will get.

Facial recognition technology pros

Facial recognition technology is prolific, and its popularity is due to its demonstrable business advantages, including:

1. Increased security: Face recognition technology can help prevent crimes and improve safety and security. It helps to find criminals and missing people even after several years. Thanks to face recognition software, businesses are protected from theft and prevent unauthorized access into office buildings.
2. Easy integration: Most of the modern facial recognition solutions are compatible with the majority of security software. You don’t even need machine learning skills to start using them.
3. Efficiency: Previously, the identification process was carried out manually by a person. It took a lot of time and was inefficient. With the help of face recognition technologies, you can monitor the attendance of employees or other visitors.
4. Easy management: Systems based on face recognition are fully automated. That’s why managing records and keeping track of daily activities is much easier.

As it has been pointed out, face recognition provides multiple opportunities for businesses and organizations. Although it still has some inaccuracies and imperfections, its capabilities are likely to expand.

In order to see how face recognition technology is developing and what changes this market is undergoing, let’s take a glance at the global facial recognition market, its opportunities, and face recognition future trends.

Global facial recognition market

The market is mostly driven by small and large organizations that are striving for the digitalization of their mundane tasks and increased capacity for employee creativity. The facial recognition market is worth approximately $5 billion, and that is expected to double by 2025. New data identifies 3D and 2D facial recognition as the most lucrative applications of facial recognition. Technavio’s research identifies the main actors in the facial recognition market as government, BFSI and transportation.

Technology insights

In 2022, new technologies like facial analytics and cloud-based solutions are poised to change the facial recognition technology market. Thanks to facial recognition capabilities, users can automatically find out where a face exists in a video or image as well as its attributes. For instance, CompreFace can analyze facial expressions including age and gender and detect the location of facial features in an image. Amazon Rekognition is capable of identifying visual geometry, mood, hair color and more. These features allow for high accuracy and effectiveness, which is why it’s often used in retail, police, office, education and healthcare settings.

The other most popular segments are access control and security/surveillance. Face recognition access control is a touchless experience that allows users to look at the face recognition device and unlock a door or phone. Security and surveillance systems are used in highly-controlled areas. Airports deploy face recognition systems at security checkpoints, and law enforcement agencies use this technology to uncover criminals’ faces or find missing people.

The retail and ecommerce sector has been actively adapting face recognition technology as well. Previously, customers had to pay bills by QR code, credit card or cash, but now they can simply scan their faces on smart devices. This allows for a more secure and user-friendly payment system.

Government agencies have been also exploiting this technology in finance and banking for digital access or cybersecurity. Some of them use biometric systems for physical security, such as monitoring access to their facilities or generating leads in criminal investigations.

Key facial recognition market segments

According to Allied Market Research, market segments can be broken down by a variety of criteria:

• By application
  • Homeland security
  • Criminal investigation
  • ID management
  • Physical security
  • Intelligent signage
  • Web application
  • Business intelligence
  • Photo indexing and sorting
  • Other (VIP recognition, automotive, and phone, PC & banking login)
• By component
  • Hardware, scanners, cameras, handheld devices, integrated devices
  • Software
• By technology
  • 3D facial recognition
  • 2D facial recognition
  • Facial analytics       

Facial recognition technology by country

A new report by Comparitech ranked countries by use of facial recognition technology. The top 10 countries with the most widespread use of facial recognition technology are:

1. China: The government and police use this technology extensively and often with invasive surveillance tactics. For instance, the city of Suzhou used the technology to publicly shame seven people who left their homes in their pajamas.
2. Russia: It’s no surprise that Russia is among the top players in the global facial recognition market. With facial recognition evident in all of the categories we covered, Russia is also turning toward facial recognition in many different areas.
3. The United Arab Emirates: People in the UAE use facial recognition to gain access to government services and register attendance at schools and work. The police in Abu Dhabi have also included face recognition technology in their patrol cars to help them identify suspicious people.
4. Japan, India, Chile: All of these countries use facial recognition technology to different degrees. For example, Japan uses facial recognition alongside citizens’ social media accounts to track down criminals. In Chile, the majority of citizens will have electronic identity cards that use facial recognition by 2022. In India, there are around 16 different face recognition systems in central and state governments.
5. Australia and Brazil: In Australia, facial recognition technology is actively used by the police, while in Brazil it’s used in schools and public transport.
6. Argentina: Face recognition technology is used to detect juvenile suspects, track attendance in schools, and facilitate the payment process in public transport.
7. France, Hungary, Malaysia and the United Kingdom: All four of these countries use facial recognition technologies for every category that we mentioned above. They have widespread or growing use of face recognition in government, police, banking and airports, and its use is also growing across public transport systems.
8. Mexico and the United States: There’s growing or widespread use of face recognition technology in all areas.
9. Romania, Spain and Taiwan: Each of these countries has increased use of facial recognition technology in most areas but all have different fields that remain untouched at present. For instance, Romanian public transport systems, Spanish schools, and Taiwanese buses don’t have face recognition technology.
10. Kazakhstan, Sweden, Thailand and South Africa: These countries have varying degrees of face recognition technology in each category. But in Sweden, facial recognition technology has been banned in schools.

Uses of facial recognition technology

Today, face recognition technology is used in several major industries and primarily as a means of personal identification. The most common use cases of facial recognition technology are:

• Phone unlocking: Face recognition is one of the simplest ways to unlock your phone. Once the phone scans the user’s face and stores the data, its inbuilt camera uses face recognition to scan a face and unlock the phone. This method is quicker and more efficient than other biometric means like tracing patterns, entering passwords, etc.
• Crime prevention: Face recognition is instrumental in detecting known burglars and shoplifters in public spaces. The system comprises a comprehensive database of criminals and matches new information to spot them within seconds. This technology has helped crime and forensic branches prevent many crimes to date.
• Attendance management: Face recognition is also being used to monitor employee attendance at the entrances of major companies. The main advantage is that it simplifies the employee experience and simultaneously solves the problem of tracking and tallying attendance information.
• Assistance to the visually impaired: Today, face recognition apps are being used to help visually impaired people detect the emotions of other people they interact with. Their phone uses specific vibrations to alert a person about others’ current emotional states based on their facial expressions.

Facial recognition trends

We’ve already touched upon some main market insights and the most popular areas where facial recognition systems are used. Now let’s look at the most popular face recognition trends in 2022.

1. Increased accuracy

Despite many calls to ban facial recognition applications, the technology continues to develop and become more and more accurate. Some evolutions from the past few years include better images with better cameras, 3D face recognition, more accurate face recognition, neural network algorithms, on-chip processing and edge computing.

Face recognition has undergone an industrial revolution. Now algorithms are increasingly tolerant of low-quality images. In NIST's 2020 tests, the best facial identification algorithm has an error rate of 0.08% - that's less than one error for 1,000 images. Taking wild photography (low-quality nature pictures) as a source, the algorithms for facial recognition are more accurate today. Taking the value (0.028) from September 2020 and comparing it to the value (0.134) from June 2018, we see the algorithms are 4 times more accurate than they were two years ago. The most recent results show the system did not recognize 0.28 cases out of 10,000 on average, which is 4.6 times more accurate than two years ago. In 2018, it was 1.3 cases out of 10,000.

2. Simplification

Technological advances do not exactly imply complicated mechanisms and principles. Rather, the more advanced and complex face recognition systems become, the more user-friendly the repositories of such solutions are. For example, in CompreFace — a free and open-source face recognition solution developed by Exadel – everything is set up to launch the program easily. You don’t need to be a machine learning expert to keep it going; basic programming skills will be enough.

3. A bias problem

I’ve already touched upon this problem in a recent interview for the Hackernoon community, but let’s discuss it here. The problem is that no facial detection system is perfect. Racial discrimination in face recognition systems has always been a stumbling block. If you look at the description of many facial recognition solutions, you will see that the first feature many companies boast of is a high accuracy (over 90%). However, this is not always the outcome. A substantial range of studies proves that systems show the poorest accuracy when detecting women, people of color, and 18-30 year olds. These racially-biased algorithms can create a negative impact on people of color. For example, if police departments apply face recognition technology to identify suspects, one inaccurate result can lead to a wrongful arrest, detention, or even worse.

The solution to this problem lies in the nature of how facial recognition algorithms “learn.” They do it after being shown millions of images of human faces. So if the faces are mostly white men, the system will have difficulties in recognizing anyone else. The way out is to expand the database with faces and include people of all races and ethnicities. This is particularly true for companies that provide face recognition services and applications.

4. Masked-face recognition

The increased use of face masks due to the COVID-19 pandemic has added a layer of complexity to face recognition, but the industry has responded with more innovation. Some studies show that the accuracy of masked-face recognition can reach 99%. For example, Japan’s NEC Corp has recently launched a facial recognition system that it says takes less than one second and has an accuracy rate of more than 99.9%. Their system can be used at security gates, in office buildings and at other facilities.

How does it work? First, the system determines when a person is wearing a mask and then focuses on the uncovered parts of a face to verify the person’s identity. Such software is trained on two sets of images: one to teach the algorithm how to recognize (face detection) and a second to recognize a mask on a face (mask recognition). Currently, this recognition software is being used in multiple settings in the United States and Europe. Restaurants and hotels are using it to monitor that staff are wearing masks. Airports have started testing the technology on-site as well.

5. Face recognition in retail

Another trend is the ubiquitous popularity of facial recognition systems in retail stores. Disney began testing it in the Magic Kingdom theme park. The system captured guests’ faces and converted them into a unique number. This number was associated with the form of admission used in the park. This experiment was temporary, and guests were not required to participate. However, Disney fans had mixed feelings about such innovations. Some of them feared that their faces and secret numbers would be further used for commercial purposes, but others were excited by the new tech.

Disney is only one example of using face recognition systems in retail stores. Hundreds of shops have already installed such systems and use them in multiple ways:

• Security: Facial recognition technology helps retailers identify known shoplifters when they enter the store. Then the system sends out an alert, and an employee can come to the identified person and offer customer service or otherwise ensure the security of the store.
• Loyalty programs: Retailers use data about their program members for personalized marketing. With facial recognition in-store, it’s easier to make special offers to customers. This could include providing a returning customer with a customized display of their favorite clothes on a screen when they go up to make an order.
• Employee tracking: A face recognition system is used to replace traditional attendance systems and decrease time spent on employee attendance tracking.
• Enhanced customer service: If the face recognition system is integrated with other systems, it can provide valuable information on individual customers. This information can include data on how frequently a customer visits the store, what purchase they made last time and what they usually buy. With this information, shop assistants can provide more personalized service to every customer.
• Self-service checkout: Face recognition can be used as part of payment verification during checkout. This helps prevent identity fraud and ensures compliance with customers’ authentication requirements.

Although many retail stores take advantage of face recognition technology, there are some negative aspects. With cameras tracking behavior, movement and emotions, there are serious data privacy concerns. And, from a psychological point of view, it’s difficult to feel that your every move is recorded. These concerns are even more serious when people are being monitored without their knowledge. This is non-transparent and violates people’s privacy, putting people and their personal information in danger. One solution to this problem is to ensure opt-in consent requirements that would prevent retailers from randomly scanning faces.

6. Payment

Payment by facial recognition is no longer a futuristic dream. The entire payment process takes less than seven seconds and is completely contactless. Also, you don’t need to carry your mobile phone or bank card with you or enter a pin number – all you need to do is to scan your face. Unlike passwords that can be easily generated and cracked, your face is the only key to getting access to your bank accounts or carrying out transactions. This method is also considered less intrusive than paying with mobile phones because modern smartphones can track your location via GPS. Face recognition systems are usually so advanced that they are difficult to trick.

Face recognition in the banking industry is not limited to payment. It can also be used instead of PINs to withdraw money. First, the user should insert their bank card and have a photo taken by the ATM camera. Then they’ll confirm a password and the image will be registered in the system to complete the verification process.

7. Driver monitoring

Another popular trend is the use of facial recognition in cars. One of the major causes of car accidents in the world is related to fatigue. Facial recognition is slowly changing that. Now facial recognition systems are used to continually check a driver’s alertness on long distances. If a driver appears to be nodding off, the system can automatically slow the vehicle gradually and give an audio alert to the driver. Monitoring a driver’s facial movements can also tell the software if they are calm or angry. Depending on the situation, the software could increase the gap between the cars.

Unlocking your car with your face is a new and reliable way to reduce theft. Car owners can also set up restrictions or permissions for other family members. This way they will prevent their small children from trying to drive, or if an authorized person enters the car, the system can block the car from starting.

Face recognition can also be used to adjust car settings to a specific person. For example, it can automatically play their favorite radio station or adjust the seat position and heating. Passengers may also enjoy the benefits of face recognition. Based on the number of riders, the system can start the air conditioning in the back seat.

8. Face recognition in the cryptocurrency world

Cryptocurrencies are a revolutionary technology that skyrocketed in recent years. However, to mitigate their risks, many companies turn to facial recognition. Binance, a well-known name in cryptocurrency, has been actively using face recognition. The Binance verification process is divided into several levels — Verification, Verification Plus and Enterprise Verification. In order to do Binance verification, you upload some documents and a photo of your face. Specialized software and experienced staff will verify your documents with the information you provide, which includes:

• Personal identification: This is done in the form of official government-issued identification documents, like a passport, a national ID or driver’s license. All you need to do is to take a photo of the document.
• Personal photo: You should upload a photo of your face for processing. If your device has a camera, you can take a photo of yourself. Then, facial recognition software will use your device’s camera to match your identity with the photo you sent. You can also use the Binance application to scan a QR code to complete this step.
• Proof of address: If you want to enjoy the benefits of Verification Plus, you will need to upload a photo of either a bank statement or a bill showing your address.

Although such software in the cryptocurrency world faces worldwide scrutiny, it is still used by many companies.

9. Content moderation

Face recognition technology works well with social media, because websites can quickly identify images that are violent or inappropriate on their pages. For example, Tumblr, one of the social media giants, banned many categories of adult content on their website including photos, videos and illustrations that depict pornography. Facial recognition ensures such circumstances are less likely to arise in the future. Thanks to image recognition capabilities, the software can determine the graphic nature of photos and estimate what age and gender the subjects are.

Another face recognition feature for social media was implemented by Facebook a few years ago. The facial recognition software automatically identified people in users’ digital photo albums and offered users “tag” them all with a click, linking their accounts to the images. Facebook built one of the largest repositories of digital photos in the world (although this system was later shut down due to public scrutiny).

The future of facial recognition

With the evolution of facial recognition technologies and their constant improvements, these systems are likely to become a part of our everyday life. We should think about how this technology will affect personal privacy and how to make it less intrusive. If professional organizations, corporations and governments continue to carry information about people’s facial data, it’s important to implement consent requirements for people not to be taken by surprise.

It’s difficult to say whether this technology will gain traction from the majority of our society, but the advantages that these services can result in are very real. Only time will tell how it will empower people and businesses.