Intriguing and Captivating Face Verification Technologies
Technological advancements in biometric authentication services are leading us towards the most advanced and innovative facial recognition services.
The rising trend of such services results in a surprising reaction from merchants and end-users in 2020 and can also be predicted in 2021.
In this era of digitization and automation, the cutting-edge technology of facial verification technology is reshaping the landscape technological era in the upcoming years.
Some predictive trends are:
- Latest innovative technologies and trendsetters
- Cutting edge algorithms of artificial intelligence
- Market-dominant use-cases from 2019 to 2024
- Facial recognition technology in various regions: China, the United States, India, United Kingdom, Brazil, Russia, European Union …
- Security vs. Privacy: two sides of the same coin.
- Can facial verification technology be tricked?
- A movement towards AI-powered hybrid solutions
Let’s have a deep dive.
How Does Face Verification Actually Work?
Facial recognition technology involves the processes of verification of customers using their faces to deter fraudulent activities. Facial features are closely analyzed, captured, and compared for enhancing the security features
- The process of face detection is a crucial step that involves the detection of the face of the customer present in image and video.
- The process of face capturing involves the conversion of face analog information into digital information based on facial features.
- The face matching process ensures that the face belongs to the same person under detection.
In today’s generation, identity verification using biometric authentication is acquired by many financial infrastructures for fraud detection and customer due diligence.
We all are uniquely identified by our fingerprint, iris, DNA, and of course facial features.
Before having a deep insight, let’s get to know ourselves about these two keywords: “authentication” and “identification”.
Facial Recognition Technology For User Authentication and Identification
To uniquely identify and verify customers, biometrics is used involving a set of verifiable and recognizable data that uniquely belongs to the specific person.
For a better insight into biometric definitions, visit the following site to acknowledge yourself.
Authentication: Is the answer to the question “are you actually who are what you say or claim to be?”
Identification: Is the answer to the question “who are you?”
Following are some examples of facial verification and biometric authentication:
- 2D or 3D sensors are incorporated in numerous financial infrastructures for the purpose of biometric authentication. Using innovative artificial intelligence algorithms, the captured information is transformed into digital data and then converted information is matched against the user information stored in the database.
- Biometric systems based on automated features can be used for the purpose of unique identification and verification of users. Those automated systems involve checking someone’s identity using facial features such as nose bridge, lip contour, the shape of ears, iris scan, voice recognition, eye spacing, and so on.
This can also be implemented in an unstable environment or in the middle of the crowd. It is proved from the performance achieved by TWB’s Live Face Identification System (LFIS), an advanced solution resulting from our long-standing proficiency in biometrics.
- Facial recognition technology is also incorporated by the Owner of the iPhone X to deter fraudulent activities. However, in late 2017, China heavily criticized Apple’s Face ID Biometric Authentication because this feature was unable to differentiate between Chinese faces.
Different characteristics of the human body such as human fingerprint, iris scan, voice recognition, digitization of veins in palm and face, and also behavioral and etiquette measurements.
Why Should You Choose to Face Verification?
AI-powered facial authentication is the most preferable method of biometric verification acquired by numerous organizations to detect fraud.
The main reason for the adoption of facial authentication service is the service is convenient, easy to deploy, accurate, and efficient.
Facial authentication does not require any physical contact between end-to-end users.
The face matching process is for the purpose of user authentication and can be carried out within a few seconds frictionlessly with a high precision rate.
#1 Latest Facial Recognition Technologies
Technologies are running in a race against each other for biometric innovations powered by artificial intelligence and machine learning.
Significant organizations such as Apple, Facebook, Google, Amazon, Microsoft are very much evolving with the latest facial recognition technologies to fight a battle against fraudulent activities.
Organizations, as well as web giants, are now regularly publishing their latest discoveries based on deep learning, artificial intelligence, machine learning, face analysis, online identity verification for the better acknowledgment of their readers and further their knowledge.
Let’s jump right in.
Facebook and Google
Facebook announced the deep face program in 2014, which determines whether or not 2 or more photos belong to the same end-user under observation or not, with a high rate of accuracy up to 97.25% frictionlessly within a few seconds.
Moreover, enhancement was seen in FaceNet by Google in June 2015. FaceNet achieved a new record of accuracy up to 99.63% i.e. (0.9963 ± 0.0009) on the widely used Labeled Faces in the Wild (LFW) dataset.
With the utilization of new algorithms of an artificial neural network, one of the companies from Mountain View has managed to link a face to its owner with almost perfect results.
The technology of facial recognition is also incorporated in Google Photos, which is used to sort images and after that automatically tag people based on the recognition of a person. As the cutting-edge technology of biometric authentication is proving to be significantly important, it was quickly followed by the online release of an unofficial open-source version of OpenFace.
According to the report from Ars Technica, Amazon is actively promoting cloud-based face verification services in May 2018. This service is also called Rekognition which is proving to be mandatorily enforced by law enforcement agencies. Face verification solutions can recognize almost 100 people in a single image and can perform a face matching process against a repository that contains millions and billions of faces.
According to the report from Newsweek, Amazon’s face verification solutions falsely identified 28 members of the United States Congress as criminals or fraudsters.
In 2014, the well-known algorithm called GaussianFace discovered by the researchers of The Chinese University of Hong Kong achieved the score of face verification of 98.52% as compared to humans i.e. 97.53%. A magnificent rating, despite weaknesses regarding memory amplitude required and calculation times.
Microsoft, IBM and Megvii
In February 2018, a study conducted by MIT researchers ensured that IBM, Microsoft, and china-based Megvii tools (FACE++) had high error rates while the comparison of an identification of dark-skin women and light-skin men.
During the second half of 2018, it was announced by Microsoft in a blog post that they have considerably enhanced facial verification technology.
Biometric Matching Technology Providers
At the end of May 2018, the United States Homeland Security Science and Technology Directorate published the results of sponsored tests at the Maryland Test Facility, abbreviated as MdTF, which were done in March. These tests precisely analyze 12 well known online facial verification services in a corridor measure of 2m by 2.5m.
Facial verification solutions utilized by Thales’s solutions achieved the acquisition rate of 99.44 percent against the average of 68 percent within a few seconds. A Vendor True verification Rate of 98 percent within 5 seconds as compared to an average of 66 percent. The error rate also reduced from 32 percent to 1 percent.
Enhanced Performance standards: In November 2018, it was published by the National Institute of Standards and Technology (NISA) claims that verification accuracy for 127 algorithms with the association of performance with the names of participants.
Additional results were provided by the National Institute of Standards and Technology (NIST) Ongoing Face Verification Vendor Test (FRVT) which are mentioned in the following report.
On January 2020, It was also demonstrated by the National Institute of Standards and Technology (NIST) that the enhanced online facial recognition systems are now biased regarding sex, color, or race by ITIF which proved the critics wrong.
“In the middle of June 2020, IBM denied offering AI-based face verification technologies, also the development and research activities came to an end by IBM. Moreover, the United States’s Law enforcement agency jerked facial recognition technologies of the well-known organization, Microsoft. ”
“On June 10th, it was published in a blogpost that a one year prohibition was adopted by Amazon on the use of its automation and technology by police. This trendy eCommerce platform claims that we are giving time to federal laws and regulations to acquire measures to be initialized and protect human rights and civil libraries which are included in this domain. ”
Facial Recognition and Emotion Detection
For real-time and statistical images, identification and recognition of facial expressions is the process that involves mapping of face expressions depicting human emotions such as joy, sadness, fear, surprise, happiness, sorrow, anger, disgust, etc.
Wide-range of organizations are acquired applications that involve facial recognition technologies and automation processes.
There’s a difference between facial recognition systems and emotion detection technologies. The main purpose of facial recognition systems is to identify facial features whereas emotion detection involves the detection of expressions.
Appearance and analysis of the unique geometric facial features are represented by facial expressions. Parameters are extracted using OCR data extraction technology from the transformed images such as dynamic models, eigenfaces, and 3D models.
Providers of such technology include Sightcorp, Affectiva, Noldus, Kairos (emotion and face detection and recognition for brand marketing).
#2 Jumping In Into Deep Learning
The evolving technologies that are acquired by a wide range of organizations and financial infrastructures are backed by algorithms based on artificial intelligence, machine learning, and more precise and accurate, deep learning, which are powering automation systems. Because of machine learning and deep learning algorithms, systems are getting smarter enough to memorize data.
But the question arises why is this significant?
The Latest- generated algorithms that were developed by Thales and other key players revolve around the concept of emerging artificial intelligence and deep learning.
But what’s the result?
Online face verification systems are enhanced with every passing day backed by Artificial intelligence, machine learning, and deep learning-based algorithms.
It is recently reported by NIST that in the last five years, immense gain in the accuracy and efficiency of face verification technologies has exceeded improvement in the upcoming time period.
Advancement and enhancement are seen in facial recognition algorithms in 2018 as compared to 2013.
NIST discovered that while the image matching process, failure rate decreased from 4% to 0.2% from 2014 to 2018 in a database possessing 26.6 million images.
You absolutely read it right!
20x times improvement was seen within the tenure of 4 years.
Let’s think about it this way:
Facial recognition systems are enhanced and are backed with high rates of accuracy and efficiency with the help of artificial neural network algorithms.
Let’s have a look at this video to find out more:
#3 Market Trends Of Facial Recognition
Trends of Facial Recognition
According to the study published in June 2019 predicts that the global revenue market of facial recognition technology successfully generated $7billion by 2024, which was supported by a compound annual growth rate (CAGR) of 16 percent within the tenure of 2019 – 2024.
The market estimates $3.2 billion by the year 2019.
A significant role was played by two important drivers for the immense growth in facial recognition technology in public sectors in numerous applications in reshaping the landscape of AI-powered facial recognition technologies.
The most well-known facial recognition vendors include the following:
Accenture, NEC, Nuance, Aware, BioID, Certibio, Thales, HYPR, Idemia, Leidos, M2SYS, Phonexia, Fujitsu, Fulcrum Biometrics and Smilepass
The main applications of facial recognition technologies can be categorized into the following categories.
What Is The Purpose Of Face Verification Technologies?
Following are there well-known categories where AI-powered facial recognition technologies are being used:
Security Law Enforcement By Higher Authorities
This market trend faced immense growth actively to deter fraudulent activities and financial terrorism.
There are evident advantages of facial recognition systems incorporated in financial infrastructures to combat fraudulent activities and illicit money transfer.
- Face verification services play a crucial role while the process of document authentication and verification of customers, which is often combined with biometric technologies such as fingerprint, iris scan, voice recognition, behavioral measurements for the prevention of ID fraud and identity theft.
- Facial authentication is used at border checks for the comparison of an image on a digitized biometric passport with the face of an individual. At Roissy Charles de Gaulle airport in Paris, TBW was responsible for supplying new automated features and services for the PARAFE system (Automated Fast Track Crossing at External Borders). During the year 2018, this solution has been conceived to facilitate the evolution from fingerprint recognition to face verification.
- Facial biometrics are also acquired by police checks. Its use is strictly controlled by Europe. In 2016, the one who was responsible for the Brussels terror attack, “The man in the hat” was successfully identified and verified with the help of FBI facial verification systems. In 2017, UEFA Champions League Final was secured with the implementation of facial verification technologies by the South Wales Police.
- The Federal Bureau of Investigation (FBI) has access to the driving license photos of almost 18 states. 26 states, or probably up to 30 states in the United States allow the law enforcement agencies to run searches against their repositories of official documents such as driving license and ID photos.
- Aerial cameras combined with drones provide an interesting duo for enhanced face verification services which is acquired for security purposes in massive gatherings. In June 2018, according to the Keesing Journal of Documents and Identity, some drones are efficient enough to carry the weighted lens up to 10-kilo that can identify the criminal from the height of almost 100 meters to 800 meters. Drones have an unlimited power supply as they can be connected to the ground via a power cable. Communication cannot be intercepted with the involvement of a power cable.
- CCTV systems based on facial recognition technologies can improve the performance of security measures to combat fraudulent activities.
The usage of facial recognition technology can be seen in the following four examples:
- Identification and tracking of criminals
- Identification and suspecting exploit child
- Acceleration and supporting of the investigation process
- Finding disoriented adults and missing children
Identification and tracking of criminals
CCTV systems based on face verification enable police stations and other investigating sectors to accelerate the examination process based on a past criminal record. Preventive actions can be adopted by police stations. By utilizing the image of known criminals from external images or a video or an image from a database. This way operators can react to the situation before it’s too late.
Identification and suspecting exploit child
The process of isolating an individual from a sequence of videos is very time-taking and critical. Face verification technologies can accelerate the operations performed by investigators in finding the exploited child.
Video analytics plays a crucial role in tracking the activity on a map, building chronologies, discovering non-obvious connections, revealing details among the players in a case.
Acceleration and supporting of the investigation process
CCTV systems involving facial recognition technologies can accelerate the process of investigation using video evidence after the incident.
Investigators can definitely develop an understanding of the situation using the information gathered from the video sequences and accelerate the process of investigation.
Finding disoriented adults and missing children
Investigation processes can be accelerated by enabling facial recognition CCTV systems. Operational efforts are accelerated by matching the image of missing children with the past appearances of the face captured in a video. Police checks can use facial recognition technologies to search video sequences i.e. video analytics of the estimated location and estimated time when the child was declared to be missing.
In that way, police officers can continue their investigation by utilizing the information acquired from the video from the suspected location during the claimed time where the child was seen for the very last time. Alert or warning can be generated whenever there is a successful match. Police can then confirm the gathered information and continue the examination of the process which is necessary to find the missing child. The same process is applicable in order to find disoriented adults, for example, amnesia, dementia, epilepsy, or Alzheimer’s disease.
Technological advancements can be seen in the health sector to fight a strong battle against fraudulent activities.
Thankfully face analysis and deep learning allow us to:
- Keep track of patient’s utilization of medicine with a high rate of accuracy and efficiency.
- Detection of genetic diseases such as DiGeorge syndrome with an accuracy rate of up to 96.6 percent.
- Reassuring pain management procedures.
3. Merchandise and Retail
This is the least expected area where facial recognition technology can be adopted for the purpose of user authentication. It is the most promising technology to combat chargebacks and peps. Know Your Customer (KYC) was the most evolving topic in 2020.
This significant trend is being merged with the latest advancements in the technological era for enhancing customer experience.
Customer due diligence is enhanced by analyzing the shopper’s behavior. This can be done by the placement of cameras in retail stores and shopping centers.
Recently Facebook designed a system in which sales staff are provided with the information of customers which is acquired from social media profiles of individuals to expertly produce custom responses.
How long before the procedure of selfie payment?
In Hangzhou, China, facial recognition technologies and solutions have been acquired by the American king of fried chicken KFC, Chinese retail and tech giant Alibaba since the year of 2017.
#4 Depiction Of New Users
The largest market of facial recognition technologies is offered by the United States, but the massive growth in the following sector can be seen in the area of Asia Pacific region. The leading countries in the race of technologies of facial recognition are India and China.
Face verification solutions in China
Facial recognition technologies and solutions are the most evolving and trendy topic in numerous sectors in China. Those sectors may include health sectors, police, airports, financial sectors, etc.
Face verification sunglass program is growing by high regulatory authorities these days as police sectors are using it in Beijing’s outskirts.
Video monitoring is enforced all across the country in China for the empowerment of security and privacy features.
200 million monitoring cameras were used during the second half of 2018 and it is predicted that 626 million cameras will be incorporated in 2020. Chinese cities are symbolic representations of the incorporation of facial recognition towers.
Chinese governments are developing social credit systems that are enhanced by the incorporation of AI-powered face recognition technologies
The top 10 leading Chinese cities which are incorporating surveillance cameras to deter fraud include Ji’nan, Tianjin, Shanghai, Shenzhen, and Chongqing.
According to the Guardian of 2 December 2019, London is #6 and Atlanta #10.
There is more to know.
According to the New York Times published on April 14, 2019, Chinese police sectors are integrating themselves with companies that are based on artificial intelligence such as CloudWalk, SenseTime, Yitu, Megvii (in partnership with Huawei).
China’s missions for the incorporation of AI and ML-powered facial recognition technologies are enormous. The aim of the country is to become the world’s ruler in artificial intelligence by the year 2030.
In a report dated November 2, 2020, a paradox was evidenced by the privacy expert Emmanuel Pernot- Leplay that China enhances government access to personal data and also provides unbelievable biometric data protection against private entities.
Face Verification in Asia
The most significant topic will face verification solutions for the 2020 Olympic Games in Tokyo, which is postponed to September 2021. The technology of facial recognition will be used for the verification and identification of an individual to combat chargebacks and fraudulent activities. Also, the incorporation of such solutions is to provide customer due diligence, safety, and security measures, with high rates of accuracy and efficiency.
Face verification technology is undergoing trails in airports in Sydney for a much faster and safer verification of individuals.
The aadhaar project is the largest project database in India all around the globe. As of December 2020, it has already provided a digital identity verification service number of 1.27 billion residents. The authority in charge of The Unique Identification Authority of India (UIDAI) announced in September 2018 that facial verification will be launched in a phased roll-out. Facial authentication service will be merged with other user authentication services such as fingerprint, iris scan, OTP, for the safety and security of customer,
India is thriving to achieve a well-reputed position in facial recognition solutions in 2021.
The national crime records bureau (NCRB) has launched RFP inviting Business Intelligence Development Studio (BIDS) to develop nationwide facial recognition automated systems. According to the documentation of 160 pages, the system will be a centralized web application hosted in Delhi by The National Crime Records Bureau (NCRB) data center. It will also be accessible by police stations.
With the help of AI-powered face verification technologies, the identification of individuals from a CCTV video sequence or image would be much simpler and easier. Operations performed by Bureau states to catch criminals are much more accelerated with the help of biometric technologies and face verification solutions.
Some More Enormous Projects
Nationwide biometric data collection project involves the incorporation of the Superior Electoral Court (Tribunal Superior Eleitoral) in Brazil. The agenda is to create a biometric database and unique ID cards by 2020, possessing the information of 140 million citizens.
Many regions such as Gabon, Africa, Burkina Faso, Cameroon have acquired TBW to meet the challenges of biometric identity for in particular the unique identification of voters.
A countrywide program has been deployed since 2017 by Russia’s Central Bank which was designed to collect faces, iris scans, voices, and fingerprints of individuals. But according to the Biometricupdate website of 13 March 2019, Unfortunately, the process is progressing very slowly.
The initiation started in January 2020.
By the end of 2019, Moscow claims to be the world’s largest network possessing 160,000 monitoring cameras to fit correctly in the sector of face verification technologies for customer safety and security.
Russian law does not regulate unconventional face analysis and detection.
#5 Face Verification Enhancing Legal Systems
Societal and Ethical and social Challenges that were previously faced by data protection sectors were reduced tremendously with the help of facial verification technologies.
The question arises that “have these science-fiction novel worthy technological advancements have actually threatened our freedom?”
And with that our obscurity?
Biometric Protection in the United Kingdom (UK) and European Union (EU)
Regions frameworks for practicing AI-backed facial recognition technologies are provided by the General Data Protection Regulation (GDPR) in Europe and also in the United Kingdom.
Any investigation which results in Invasion of privacy of a citizen’s private life or business travel habits carries several severe penalties.
GDPR supports the principle of the Harmonized European Framework to provide clear affirmative action to protect the rights that are forgotten or neglected.
Yes, You read it right. Enforcement of one law is ensured for 500 million people.
This directive is bound to have international reverberation.
US Landscape In Biometric Authentication Technologies
After Texas and Illinois, the state of Washington in America was the third US state for the protection of biometric data through a new law introduced in June 2017.
As of January 2020, California was the fourth state.
An act was passed in June 2018 called The California Consumer Privacy Act (CCPA) which proved to be effective by 1st January 2020. The act will have a serious impact on the privacy rights and customer protection for the residents of California and for the whole nation.
The law is symbolized frequently as federal data privacy law.
CGPA has the potential to become as consequential as GDPR in that sense.
Microsoft’s president Bradford L. Smith, made a comparison between face verification technology and other products such as highly regulated medicines in July 2018. Moreover, he urged Congress to study it and oversee its use.
US Representative Alexandria Ocasio-Cortez depicted his absolute concern in the recent committee held for the purpose of facial recognition technology in May 2019, which was impactful on our civil liberties and rights.
New York law proved to be effective on 21 March 2020 which was called Stop Hacks and Improve Electronic Data Security (SHIELD).
This act is applicable to such businesses and financial infrastructures that collect the personal information of NY residents.
With that regulation, New York resides beside California.
Face Verification Bans (Somerville, Portland, San Francisco, Oakland, Portland, Boston)
Facial recognition is evolving rapidly as a law enforcement tool in civil and privacy rights. Contrary to this,on 6 May 2019, San Francisco voted to ban face verification.
It is the first ban of this kind which is imposed on facial recognition technology.
As of June 2019, San Francisco’s Board of Supervisors bars city agencies along with San Francisco PD signed the anti-surveillance ordinance from using the technology.
This absolutely includes law enforcement.
But there’s more.
On 27th June 2019, Boston Globe reported that the Somerville City Council Massachusetts voted to ban face verification, which made the city the second community to make this decision.
Foam, rinse, repeat.
- Oakland from California took the same decision and became the third US city to ban the use of facial recognition technology on 16 July 2019. An interesting aspect is that California police stations were not planning to incorporate facial verification technology and were not even planning to use it.
- In advance of the new California law, during the second half of December 2019, San Diego took the same decision to ban facial recognition technology. This new law called Assembly Bill 215 about face verification and other biometric services specifically prohibits the use of police body cameras in California.
- It is reported by Boston Herald that Boston voted to ban face verification technologies and solutions by police on 24 June 2020.
- On 9 September 2020, Portland (Oregon) decided to ban facial surveillance technology. It is the first city to extend to private entities in places of public accommodation. (CNN)
Since Portland rulings, Sommerville, Oakland, San Francisco, and now San Diego Boston, and, the debate gets louder in many cities and not only in the United States.
During the second half of August 2019, Sweden’s data protection authority decided to ban facial verification technologies in Europe in educational sectors and fined a local school, which was the first GDPR penalty in the country.
What Should We Do For The Betterment of Regulatory Technologies?
- Should other cities and states follow this example?
- Is the ban just a “pause” button for the betterment of risk assets?
- Is this a step back for public safety?
- At which level there is an existence of a policy vacuum?
As the US government is being pressured by the activists to ban the technology and to regulate the providers.
But as of January 2021, there is still no federal legal framework to address the issue.
An act is under discussion by the European Union commission to haphazard the use of face verification technology. Ursula von der Leyen, the president of the European Commission wants to synergize human and artificial intelligence with an ethical approach. She is planning to publish a blueprint based on Artificial intelligence legislation very soon.
The very first draft of the European Union commission whitepaper is thankfully available online.
It is mentioned in the document that “A time-limited ban on the use of face verification by public and private sectors in public places. ”
The questions of privacy, function creep, and consent, data collected for one purpose being used for another, are central to the debate.
You can have a deep insight into biometric data protection (the UK, US, and EU perspective) in the following data set.
Aadhaar: India’s National Identification Scheme
Puttaswamy judgment in India which was delivered on 27 August 2017, the supreme court laid down the right to privacy in the constitution of the country. This decision has rebalanced the relationship between the state and the citizen. A new challenge was also faced by Aadhaar for project expansion.
On 28 February 2019, the implementation of the Country’s biometric EID program in private sectors and entities was approved by the Indian government.
The rebound effect of this scheme is to enhance the security and privacy features in legal systems and their professions.
The incorporation of data protection officers is becoming mandatory for businesses by the ambassadors and guardians of data protection regulations.
Moreover, Aadhaar acquired anti-money laundering to make the use of Aadhaar simpler, safe, and secure. The notification was issued on Wednesday amending the anti-money laundering (AML) and Know Your Customer (KYC) compliance for the maintenance of record in 2005.
#6 Facial Recognition Hackers – The Rebels
Critical voices are still raised despite this legal and technical arsenal designed to protect citizens, data, and their anonymity.
Some parties are alarmed and concerned by these developments. And some of them have taken action.
But the question arises that “can facial recognition technology be fooled?”
- A solution was invented by Grigory Bakunovin in Russia to Escape an eye permanently watching our movement to confuse face verification devices. He successfully developed an algorithm that was capable of creating special makeup to trick the software. But he has chosen not to bring his software to the market after realizing how easily criminals could use it.
- Berlin artist Adam Harvey in Germany successfully discovered a similar device called CV Dazzle. To prevent the detection and enhancement of devices, he is currently working on clothing feature patterns.
- During the second half of 2017, a mask was successfully used by a Vietnamese company to hack the facial recognition system of Apple’s iPhone X. However the implementation of this hack is very complex on a large scale.
- During that same era, researchers from the German company discovered a hack that allows individuals to dodge facial authentication systems of Windows 10 Hello by imprinting facial images in infrared.
- In May 2018, Forbes announced in his article that the researchers from the University of Toronto discovered an algorithm for the disruption of face verification technologies aka privacy filters.
- A cloaking app called Fawkes was detailed by the verge in August 2020. This software is capable of distorting selfies and pictures of the individuals which you might leave on social media platforms. The tool was discovered from the University of Chicago’s Sand Lab.
“It all boils down to the fact that an individual can use filters to modify specific pixels in a picture before uploading it into the website. These modifications can easily dodge and fool facial recognition algorithms. ”
- A tool called Anonymer was made available by Generated Media in November 2020. This software is capable of generating a series of synthetic portraits from an image that can be uploaded by an individual. Statistically, images will look like the face of the user but will fool the software according to the TNW website. It provides interesting solutions to manipulate and fool systems like Clearview AI that are discarding millions of faces from social media platforms.
“On 27 November 2020, Anonymizer was tested. But 40+ doppelgangers were far from looking like the original photo uploaded by an individual. ”
Industries and infrastructures are working on group standardization and anti-spoofing mechanisms which revolves around the following 2 topics:
- Ensuring that the image is captured from an individual, not from a photograph or video screen i.e. 2D or a mask i.e. 3D (liveness detection and liveness checks)
- Ensuring that the face image (morphed portraits) of two or more than two individuals are not joined in the reference document verification such as a passport.
#7 A Movement Towards Hybrid Solutions
Future authentication and verification solutions will possess all the possible aspects of biometric authentication to deter fraudulent activities.
This will lead this unpredictable world towards biometric mix or “biometrix” which guarantees privacy and security for all the stakeholders and in the ecosystem.
TBW Gemalto IdCloud Fraud Prevention, risk assessment, and fraud detection is playing a crucial role in financial infrastructures to fight a battle against fraudulent activities.
This solution involves the merge of keying patterns, IP-addresses of the device being used, and geolocation for the user authentication to carry out online-banking processes with safety and security measures.
The 7th trend belongs to us.
It is mandatory for us to predict enhanced solutions to secure this world by synergizing artificial intelligence with biometric services projects.
You and Facial Verification Systems
It’s your turn now to level up the game.
Upcoming years are taking us towards an unpredictable and unrecognized era of digitization.
Are you willing to fill in some openings and gaps?
You can utilize the comment section any way you want by typing in your opinions on facial recognition technology, tech, and trends, or you can post any questions related to a certain topic.
We are eagerly looking forward to hearing your perception.