Social Implications of the Facial Recognition System

Social Implications of the Facial Recognition System Leaving the password and ID documents behind, the facial recognition system has been becoming more and more ubiquitous. As fast as unlocking smart devices like phones, computers, doors and others, the facial recognition system also developed quickly that it also touches every aspect in our life, from the very personal and private matters to the way business, government and law work. We know that every technology is a double-edged knife. Some hail this technology optimistically, some belittle and criticize it, and some even reject it. This paper tries to present a reflection on how facial recognition system as one of the Internet of Things affects the society. Written by: Astrid Priscilla Dion Student Number: s2082209 Study: M-COM Teacher: Dr. Alexander van Deursen Date: 14th March 2019 SOCIAL IMPLICATIONS OF THE FACIAL RECOGNITION SYSTEM INTRODUCTION OF THE TECHNOLOGY Biometric approach such as fingerprint, iris and retina scan are considered as the most accurate ways to identify a person and therefore believed as the safest method of securing information or possession. Employees, students, participants and others prove their attendance by pressing their finger(s) to a detector. Crossing the border of two countries with eyes-scan is common to everyone. Unlocking smartphone using our fingerprint was considered as a sophisticated technology, but only for a few years since it is now has been outsmarted by the new technology, which is called as facial recognition. The idea of facial recognition has been around since a very long time ago. It got commoner with the tagging and profile identifying features in social network platforms such as Google+ and Facebook. With the rise of familiarity of this identification method, being able to unlock smartphones, tablets, computer, and other devices just by looking at them is seen as such convenience. Just give the device input by scanning your face and since then you could use your smart devices faster and safer (unless if you have a twin with identical facial characteristics). With this kind of handiness, more and more people verify themselves by facial recognition system in numerous applications such as banking service, log in to any platforms, getting inside buildings, checking in at airports or hotels, or even getting healthcare treatments. The facial recognition system is not only helpful for personal use. The institutions such as service provider and business are getting assisted too, especially with emotional/behavior analysis feature (Bobde & Deshmukh, 2014). With facial recognition surveillance, they could monitor which clients are satisfied and which clients are bored, which customers are happy and going to spend more, which customers are practical and not interested in discounts, etc. Combined with other technologies, facial recognition system could provide the marketer with the data of which customer purchase which products, when, where and how often, attracted or dispassionate by what kind of marketing, and in the end it could give the insight of the customer’s behavior. That is just one of the examples and it could already give us the acumen of how this facial recognition system works for other applications. More productivity and effectiveness could be reached at work, school, public spaces, etc. While the other physiological characteristics need to be used with the participation (furthermore, consent) of the owner, or in other words the person should be actively giving his/her biometric data, facial recognition enables a passive or non-intrusive way of collecting data (Tolba, El-Harby & El-Baz, 2005). Nowadays with modern facial recognition method, nobody needs to bother putting their finger(s) on or standing in front of the sensor to be identified. By using surveillance cameras, information of position, shape, size, distance and texture or contour of eyes, nose, cheekbones, jaw, ears, and others are captured, extracted and stored in the gallery or database. This step is known as verification which will be continued with identification, where the algorithm analyzes the data to match what it already collected at the first place (Coffin & Ingram, 1999). In the law enforcement context, the government would gather its citizens’ facial features and build a surveillance system by setting up cameras and/or drones to monitor the surroundings. So whenever someone jaywalks, rides or drives over speed, misuses public facilities, starts any crime, the person could be identified in no time and processed faster. And thus, the stronger law leads to a safer society. Although it might seem like a utopia, China and the USA have been trying to realize this approach. The US government is deploying a new facial recognition system at the southern border that would record images of people inside vehicles entering and leaving the country (to Mexico) (Levin, 2018). In China, the facial recognition system is used to identify pedestrian and jaywalkers, and even to dispense a limited amount of toilet paper in public restrooms to prevent any steals (Larson, 2018). Facial recognition system could be seen as one of the revolutionary mechanisms for a better world (or worse?). However, it is not a very pleasing concept for some people. The thought of being monitored or observed constantly worries the folk that they would have less privacy, less freedom, and less security. But is that anxiety necessary? What is to be concerned about? What changes the facial recognition system could cause? What impact would it bring to our lives? APPLICATION OF THE TECHNOLOGY The application of facial recognition system can be categorized into two: commercial application and law enforcement application. Each application possesses different implications. Commercial Application Facial recognition technology can be used in numerous consumer and business applications, but the extent of its current use in commercial settings is not completely known. Therefore, the commercial implication currently is not clear either (GAO, 2015). This paper would like to assume that at least there are two types of facial recognition application for commercial or business matters: identifying people and further, analyzing their emotions/behavior. There are some features of identifying people that are being harnessed for commercial use such as the picture or video tagging, logging in to any system or unlocking devices, targeted ads, registration and payment. Since the era of social media, people have been uploading abundant of pictures that are used by social media platforms to identify their users. Later, social media platforms could analyze their users' profile to help them tagging other users on their posts. This tagging feature could be done manually by the user who makes a post or sometimes automatically by the social media platform itself. The video or picture then is available to be viewed and/or shared by tagged users, or in some cases, their friends too, depends on their privacy setting in the social media platform. This tagging feature is expected to trigger interaction between the users whether it is online interaction or offline one. Nowadays social media like Facebook also planning to allow the users to log in with facial recognition system. When Facebook detects any suspicious activities from the users, it will ask for further verification to the user by uploading a selfie rather than just typing password (Liao, 2017). The user who wants to make new account(s), send many friend requests at the same time, set up a payment, create and edit advertising and others need to provide their facial data. This idea is meant to provide a secure and responsible environment for other Facebook users. Another case of this is when a Facebook user gets locked out of his/her account. A user then can regain access to the account by using the face to verify identity. This feature is thought to be really useful and safest option when a user cannot receive a two-factor authentication SMS or call such as when a person is traveling abroad or on a flight, or lose access to his/her email account (Constine, 2017). Although Facebook is one of the most talked topics when it comes to facial recognition login, actually Windows realized this convenience idea earlier by releasing Windows Hello. Since July 2015, any Windows 10 user with supportive hardware can log in just by looking at the screen of their laptop, computer, tablet, etc (Windows, 2015). However, Apple got more worldwide attention than Windows when it launched the Face ID in September 2017 for iPhone and iPad Pro. The True Depth camera on these products capture the facial data by projecting over 30,000 invisible dots to create a map of a user’s face and capture the infrared image of the face. Later, the image will be transformed into a mathematical representation and be compared with the registered facial data that a user enrolls in the beginning. It will also work with hats, (sun) glasses, lenses, scarves and even in the dark. This Face ID feature allows the user to unlock a device, authorize purchases in the Apple store and sign in to applications (Apple, 2018). Not every technology devices support the facial recognition system as Windows and Apple do. Luckily, there are facial recognition applications or software that can be installed by non-Windows users or non-Apple users to enjoy such luxury ease. ZoOm, Rohos Face Logon, Bio ID and others are some examples that can be used for various functions from private to office uses (BioID, 2019; FaceTec, 2019). Another example of facial recognition system is face or photo editing. Social media such as Snapchat and Instagram enable the users to put filters or stickers on their face photos. Other applications or software such as YouCam Makeup, Perfect 365, ModiFace Editor and InstaBeauty allow the users to upload their photo and do makeup editing effortlessly. Targeted advertisement feature of facial recognition system counts on the demographic aspect of facial data, especially age and gender. And how do the advertisers get these data? Some people input their own data by scanning their face, if their device supports the facial recognition system. For instance, when a 40 years old female open Facebook, she later will see ads about anti-aging skincare. When a 17 years old boy plays the football game on his tablet and is about to go to the next round, he needs to see an ad about a pair of football shoes. Another way for the advertisers to get the data is from the surveillance camera. Tesco, a British supermarket chain store is the first to implement this idea in 450 gas stations, in order to show personalized ads (Bosteels, 2013). A truck driver then will see the ads of energy drinks, instead of dry shampoo ads that should be seen by a waitress. Other features of the facial recognition system harnessed for commercial use are registration and payment. Transportation Security Administration (TSA) had laid out the plan to use facial recognition system for domestic flights check-in to speed up the process (Lee, 2018). This feature could also be used in the hospitality industry such as by hotels and tour agency and event industry. In 2018 for example, Ticket Master, a ticketing agency has started selling tickets where the buyers can opt for facial recognition to verify themselves entering an event. It triggered the public's attention when the singer Taylor Swift using this feature when for her fans to get inside and monitor the security of her concert (Knopper, 2018). In the healthcare industry, besides securing the medical facilities access, facial recognition could cut the registration time for the patients, especially in an emergency that needs treatment as soon as possible. It is an amazing reality that now some people can just smile to pay for their food. Yum China has been implementing the "Smile to Pay" facial recognition system in more than 300 restaurant chain stores across the country. Going further than just food, to support tourism and shopping, most stores in Zhejiang province in China has been also equipped with Alipay system from AliExpress, called Dragonfly (China Daily, 2019). In a similar vein, a Spanish multinational bank Banco Bilbao Vizcaya Argentaria or better known as BBVA, launched its “Selfie & Go” facial recognition payment system. In the bank’s office restaurants and cafes, the customer only needs to register their face in the first place, active their smartphone’s Bluetooth then look at the camera and the bill will be sent to the customer in no time. (BBVA, 2018). Other companies that do not want to leave behind also race with their facial recognition innovations such as Uniqul, PayPal and Diebold (Patel, 2014). Facial recognition also facilitates emotional reaction analysis through facial coding and eye movement tracking. Facial coding focuses on subterranean muscle activity going on around the eyes and mouth that reveals basic human emotional states. Combined with eye movement tracker, it is possible to provide answers for questions in retail or technology usability. Where do people look when walking down the aisle of a store? What products or offerings that attract the most attention of consumer? Where do visitors of website or application users look and for how long? More than that, the utilization of emotional analysis of facial recognition system can also identify/monitor consumer or visitor satisfaction of a product or a service. For instance, according to Oracle's Hotel 2025 report, 72% of hotel operators are expecting to deploy facial recognition system in their facilities to track the guests' satisfaction in the next four years (Xie, 2018). This idea could be implemented not only by hotels but also by other service providers from small to big scale such as restaurant, public library, banks, hospitals, government offices, retail stores, etc. Last but not least, after helping the advertisers with the data, facial recognition system can also help with evaluating the effectiveness of the advertisement by analyzing the target’s emotional reaction after seeing an ad, although this still requires further study and investigation (Lewinski, Fransen, & Tan, 2014). Law Enforcement Application Police officers and others in the law enforcement system in countries around the world have been using various technologies to help them enhance their works. However, it is inevitable that technology is also used to commit various crimes, even new and sophisticated ones. This fact intensifies the research and application for technology in law enforcement. One of the most progressive technologies is called Automated Facial Recognition (AFR) that works by analyzing facial features, generating mathematical representation of them, and then finding possible matches by comparing them to the face galleries in the database. In the United Kingdom for instance, since 2010 the rise of recorded crime, unfortunately, has been occurring with the falling number of total police workforce by almost 20 percent. Therefore the UK government embraces new technological solutions such as facial recognition system to help to boost their capability and capacity to monitor and track individuals about whom they have concerns (Davies, Dawson, & Innes, 2018). Until now, the automated facial recognition (AFR) has been tried to be applied for the several times by Metropolitan Police in London such as in Notting Hill Carnival 2016, Notting Hill Carnival 2017, Remembrance Day Service 2017, and Christmas markets in 2018. Facial recognition observation has been also successfully done by the South Wales Police in securing 2017 UEFA Champion League Final in Cardiff, although there were some concerns which will be explained later in this paper. Not in just United Kingdom, the facial recognition system has also been tried to be harnessed in the United States. Most adult Americans are already in the government's facial recognition database, resulting from the driver's license and passport photos. In 2015, at least 39 states in the US used the face recognition software with their Department of Motor Vehicles (DMV) databases to detect fraud. Databases are also found at the local level which can be more accurate and more complete. Pinellas County Sherrif’s Office in Florida may have one of the largest local face analysis databases that are searched by more than 240 agencies about 8,000 times monthly (Garvie, Bedoya, & Frankle, 2016). These law enforcement application examples would not be possible without the software from other third parties who do business in facial recognition technology. Rekognition and NEC software companies are the biggest players in this field that always try to come up with innovations, such as wearable bodycams for facial recognition (Doffman, 2019). In San Diego for instance, police officers are allowed to stop people on the street, use their smartphone or tablet to take photographs of them and try to find the country’s mugshot database, thanks to Tactical Identification System (TACIDS) (Winston, 2013). In the US, the idea of employing facial recognition system along with video surveillance cameras for law enforcement is thought to be beneficial for the public by for three reasons. First, there is no bias, profiling or discrimination done by the algorithm. The facial recognition software just simply analyzes the data and looks for “bad guy” patterns which for example showing anxiety, walking or moving aggressively, act like hiding something and others and then compare it with the database that it has. There is no racial judgment or discrimination would be made. Therefore from this respect, facial recognition system would be better law enforcer than humans. Second, it has been proven to work with a number of cases. In 2017 for instance, thanks to Amazon’s Rekognition system, the Washington County Sherrif’s Office succeeded to identify and arrest a suspect who stole more than $5,000 from local stores. A study also supports this reasoning that through comparative analysis, real-time facial recognition system could help New York Police Department deters terrorism, prevents violent crime, identifies wanted individuals, finds missing persons as well as assists in mental health situations and post-event investigation (Carter, 2018). The last reason is with facial recognition, the policing will get more efficient, more time on the streets and therefore, safer society. Studies show that having more police officers on the street leads to reduction of crimes. Facial recognition enables faster streamline identification and suspect’s process, allow the officers to get on the street again faster and reduce the crime rates (Quintas, 2018). SOCIAL IMPACT OF THE TECHNOLOGY Pitfalls in Estimation of Social Impact of Facial Recognition The application of facial recognition system for both commercial and law enforcement purposes possess some implication or consequences for the society and therefore, policy should be made to regulate them. But before we talked about it, it is necessary to discuss the four pitfalls that are commonly made in public opinion in the estimation of the social impact of technology. The Idea of Total Revolution The first pitfall is many people think that the facial recognition system will radically change our lives. We will no longer have privacy and control in the future, we do not need everything with us to pay or verify ourselves instead of our face, how strict the humans will behave since surveillance camera is everywhere, how helpless human without any facial identifier, life would be anxious that humans cannot fake their feelings with the facial recognition emotional analysis, and so on. However, instead of worrying too much, we should be reminded with how many times humans have thought like that about other technologies before. People who are trapped in this pitfall believe in technological determinism that human cannot do anything to technological influence and it will be converted into social realities much too fast. But in fact, technological innovations rarely lead to societal revolutions straight away because they need to touch the social, economic and cultural aspects to cause any revolutions (van Dijk, 2019). The Idea of Social Continuity The second pitfall is people think that new technologies are seen as mere continuous improvements of existing technologies. Facial recognition then is purely just a new more advanced biometric identification system, a continuation from fingerprint, iris and retina or voice scan. However, we could see that people who are trapped in this pitfall underestimate the transforming potential of a technology because not all changes that are brought by a technology are incremental (van Dijk, 2019). It is true that facial recognition system is not yet a revolution, but it is also a fact that it has been making transformation of the way things in the world work. Facial recognition would only need to trigger structural changes in the society to turn a transformation into a revolution. The Idea of a Technological Fix The third pitfall attracts people who believe that new technology is the ultimate solution of social problems. With facial recognition system, nobody will jaywalk, rob or steal, fight in the streets and throw the trash carelessly. Less crime, more lives are saved and a peaceful world is not a far-fetched idea. With facial recognition system, the advertising would always be effective and the cost could be cut, therefore leads to more affordable products for everyone in society. What is so wrong with this assumption? It is too superficial. The problems in society such as crimes and consumerism have deeper and more complex causes. Facial recognition system alone cannot solve or fix any problem without any economic, political or other organizational measures in other fields concurrently and simultaneously (van Dijk, 2019). Instrumentalism The last pitfall traps the one who thinks that technology can be used for any purpose, either good or bad, depends on how people use it. Facial recognition system could, of course, result in bad effects, but only if we have bad intention or purpose about it. And the other way around, facial recognition system could result in bad effects in solving the social problems, if we fix that technology to do that. Those statements are fairly true. But although technology is a tool, an instrument, we should not forget the fact that tools and purposes, means and goals, they influence each other (van Dijk, 2019). Facial recognition system is developed to assist us in reaching positive goals/purposes in the first place, but when it is released in the market for practice, it is not impossible to find other unexpected new uses of facial recognition. This new use will trigger the innovations of facial recognition technology with new characteristics that can influence new purposes (even the negative ones), and so on. Facial Recognition System as Trends Amplifier Besides the pitfalls, before we talk about the policy concerns, it is also important to take a look at trends in the society that will be reinforced by the application of a technology. This paper believes that at least there are 5 major trends that will be influenced by facial recognition system. First, the main affected trend would be the rise of registration for control. Control in the society is not possible without the registration/identification of the society itself. Since the moment a person is born, he/she should be registered in the database of civil government by having a birth certificate and Identity Card (ID card) later. A person should keep informing the government about any changes in his/her address, sex, gender, job, marital status, citizenship, abroad travel history, etc. Identity documents such as ID card, passport, driver license contain the information of a person along with the photograph and biometric data such as iris and retina measurement, fingerprints and the newest one, facial data. These biometric data helps the government to identify a person accurately and quickly in case of the absence of any identity document. But what is so different about facial recognition? As mentioned at the beginning of this paper, other biometric data requires active participation from the person to be identified by the biometric scanner. Facial data, on the other hand, can be collected without a person's consent. Any person who does any illegal acts (even the small ones) cannot refuse to be scanned by surveillance cameras and therefore cannot refuse to be identified accurately and cannot run from the consequences. For that sake of justice controlling, the need for registration will be raised, just like before when a new technology interferes the way population record works. Once facial recognition registration is implemented in a certain area or field, others will follow to have the same level of surveillance. Therefore, by reinforcing this trend, facial recognition system just made the harm of privacy as inevitable in the future. Second, the facial recognition system reinforces the space trend by increasing the mobility. Regarding the ease or quickness of registration enabled by the facial recognition system, the mobility of the population will be increased too. Shorter line and time for queuing and checking in. Mobility also refers to the movement of things and ideas. Facial recognition system facilitates faster logistic operation of products and money circulation, which escort us to increased economic mobility. And this economic mobility leads to the third trend related to capitalism, the growth of instability. Facial recognition reinforces the speed of change in the economy. The personalized or individualized advertisement boosts the ever faster selling and buying in the market. Hence, it confuses the product's life cycle and triggers the yo-yo movement on the stock market. This complicates government and regulatory reaction in case of problems. Fortunately, government reaction was faster than ever before in the current crisis (van Dijk, 2019). In relation to the mobility of idea, facial recognition reinforces the fourth trend, the growth of social inequality. There will be a gap between the societies that employ the technology and the ones that do not. For example, inequality will be found between the agencies (such as police forces or government departments) that gain access to facial recognition database and other agencies that do not. Inequality will also grow between people who master digital skills and people who do not. There are four types of digital skills which are operational skill, formal skill, information skill and strategic skills. And the differential level of each of digital skill type will just strengthen the social inequality. And the last trend that will be bolstered by facial recognition system is the rise of public participation. The concern of facial recognition application will foster discussion in the society, and obligate people to contribute to any law enforcement actions. FURTHER CONCERN OF TECHNOLOGY As we have reviewed how facial recognition impacts our current life—and how it possibly will impact our lives in the future—further concern of this technology that is mainly discussed is related to the policy. When it comes to policies, there are three major issues that should be taken into account seriously: privacy, security and accuracy. When we go out from home and head to the supermarket, we know that we are going to be recorded by security cameras and we are fine with it. We have gotten used to the idea of being photographed/recorded or monitored constantly. So then why are some people getting uncomfortable and alarmed about their privacy right now? It is because we lost the anonymity of our action. And this is just actually just the surface of privacy concern. There are three levels of facial recognition impact on privacy. Level I, the least privacy-intrusive form of facial recognition is called individual counting where people’s facial information is gathered on an aggregate basis but not used for tailoring advertisement or messages. No retained information links to individuals or their property. This happens with the surveillance cameras that only track gazes or record passerby demographics but do not store the facial images. Level II which is called Individual Targeting used the collected data for tailoring ads or other messages to the individuals. This happens for example when a consumer comes to the gas station and being greeted by the monitor as “Good Evening, Sir” and being shown to a shaving razor advertisement. Level III which is called Individual Identification links the facial information that is collected to the individuals’ identity and/or property. This happens for example when facial recognition cameras record the unique biometric data points of an individual’s face in order to pinpoint his/her image on the web or log his/her location (Center for Democracy & Technology, 2012). Although the application of facial recognition is driven by security concern, it also possesses security risks. Facial recognition technology is still at its relative infancy and therefore also vulnerable to any data leaks and misuse of its power and potentials. Security concern has also been related to accuracy since there have been cases of error. There are two concepts of error or failure in facial recognition system: 1) False negative; when the system fails to match a person's face image that is actually contained in the database. In this case, the system will show zero results responding to a query; 2) False positive; when the system matches a person's face to an image in a database, but the match is in fact, incorrect. This happens when a police officer submits a photo "John" but the database tells that it is a photo of "Jack" (Grother, Quinn, & Ngan, 2017). Research indicates that ethnic minorities, people of color, and women are more prone to be misidentified by the facial recognition system (Klare et al, 2012). The trial of facial recognition use by Metropolitan, the biggest police force in United Kingdom resulted in 98% inaccuracy by producing 104 alerts which only 2 were confirmed to be positive matches. The Metropolitan police also recorded the false-negative identification rate of averagely 25% in 2017. The South Wales Police has reported 2,400 false positives in 15 deployments since June 2017 (Campbell, 2018; Sharman, 2017). Taking the lesson from these concerns, policymaking is necessary. Although the "market" of facial recognition use is still fairly small, with business intervention it could grow wider and touches every aspect quicker than we think. Therefore, it is important for the society to call for legislation that regulates this technology before it is implemented without society's input. But as with most innovation, legislation lags significantly behind implementation. At least the European Union's General Data Protection Regulation (GDPR) set clear rules about this that facial data falls into biometric data category under sensitive data. Its interpretation is addressed by the Article 29 Working Party. In EU countries, facial data can only be collected if the system is in line with Data Protection Impact Assessment and security law, getting a person's consent and anonymize/pseudonymize the data. The exception can be made that facial data is allowed to be collected and used when it is required to carrying out employment, social security, legal issues, protecting vital interest of an individual and aid in the public interest (Center for Democracy & Technology, 2012; Whitener & Aragon, 2019). With the always-changing situation (it has been discussed before it triggers instability in the society), it cannot be said that the European Union has the perfect legislation for this case. Improvement should be continuously made, especially by the countries that have not had any clear road map about facial recognition system application. Therefore, by reviewing the literature this paper would like to propose some insight into this technology's further policy concern. There are further elaborations needed regarding the performance, operation and evaluation of facial recognition system. And there are a number of questions that should be asked of any facial recognition system application, such as: - Are subjects aware that their images have been obtained for and included in the gallery database? Have they consented? In what form? Have policies on access to the gallery been thoughtfully determined and explicitly stated? Are people aware that their images are being captured for identification purposes? Have and how have they consented? Have policies on access to all information captured and generated by the system been thoughtfully determined and explicitly stated? Does the deployment of FRT in a particular context violate reasonable expectations of subjects? Have policies on the use of information captured via FRT been thoughtfully determined and explicitly stated? Is information gleaned from FRT made available to external actors and under what terms? (Introna & Nissenbaum, 2009) In relation to the security and accuracy concerns, in operational settings, facial recognition system requires highly trained and professional staff. It is important that they understand the operating tolerances and exceptions from the system and able to interpret it respectively. When it comes to the application, we should consider this technology as “assisted facial recognition system” instead of “automated facial recognition system” (Davies, Innes, & Dawson, 2018). All positive matches in the facial recognition identification in the first instance should be treated as potential false positive until verified by other overlapping and/or independent sources (Lynch, 2018). In conclusion, there have been innovations brought by facial recognition system, and there will be more in the future. It will inexorably give some implications to the society, but the implications will also depend on the society itself, owing to the fact that society and technology are influencing each other to shape the world. REFERENCES Apple. (2018). About Face ID advanced technology. Retrieved March 10, 2019, from https://support.apple.com/en-us/HT208108 BBVA. (2018). BBVA launches facial recognition payments. Retrieved March 10, 2019, from https://www.bbva.com/en/bbva-launches-facial-recognition-payments/ BioID. (2019). BioID Facial Recognition App with face login. Retrieved March 10, 2019, from https://www.bioid.com/facial-recognition-app/ Bobde, S., & Deshmukh, S. (2014). Face Recognition Technology. International Journal of Computer Science and Mobile Computing, 3(10). Bosteels, K. (2013). Tesco installing facial recognition camera’s for advertisers. Retrieved March 10, 2019, from https://www.retaildetail.eu/en/news/algemeen/tesco-installing-facial-recognitioncamera’s-advertisers Campbell, L. (2018). Why regulating facial recognition technology is so problematic-and necessary. Retrieved March 12, 2019, from https://phys.org/news/2018-11-facial-recognition-technologyproblematicand.html#nRlv Carter, A. M. (2018). Facing Reality: The Benefits and Challenges of Facial Recognition for the NYPD. Naval Post Graduate School. Center for Democracy & Technology. (2012). Facial Recognition & Privacy: An EU-US Perspective. Washington. China Daily. (2019). China’s first facial recognition payment-based shopping street opens in Wenzhou. Retrieved March 10, 2019, from http://www.chinadaily.com.cn/a/201901/18/WS5c4142c7a3106c65c34e53dd.html Coffin, J. S., & Ingram, D. (1999). 5,991,429. United States of America: U.S. Patent. Constine, J. (2017). Facebook can unlock your account with facial recognition. Retrieved March 10, 2019, from https://techcrunch.com/2017/09/29/facebook-face-id/ Davies, B., Dawson, A., & Innes, M. (2018). How facial recognition technology aids police. Retrieved March 11, 2013, from https://phys.org/news/2018-11-facial-recognition-technology-aidspolice.html Davies, B., Innes, M., & Dawson, A. (2018). An Evaluation of South Wales Police’s Use of Automated Facial Recognition. Cardiff. Doffman, Z. (2019). Facial Recognition is Coming to Police Body - Worn Cameras in 2019. Retrieved March 11, 2019, from https://www.forbes.com/sites/zakdoffman/2019/01/10/body-worn-2-0how-iot-facial-recognition-is-set-to-change-frontline-policing/ FaceTec. (2019). ZoOm - 3D Face Liveness & Authentication Software. Retrieved March 10, 2019, from https://www.zoomlogin.com GAO. (2015). Facial Recognition Technology: Commercial Uses, Privacy Issues, and Applicable Federal Law. Washington. Retrieved from https://www.gao.gov/products/GAO-15-621 Garvie, C., Bedoya, A., & Frankle, J. (2016). Unregulated Police Face Recognition in America. Georgetown Law Center on Privacty & Technology Retrieved March 11, 2019, from https://www.perpetuallineup.org/ Grother, P., Quinn, G., & Ngan, M. (2017). Face In Video Evaluation (FIVE): Face Recognition of NonCooperative Subjects. Washington. https://doi.org/10.6028/NIST.IR.8173 Introna, L., & Nissenbaum, H. (2009). Facial Recognition Technology: A Survey of Policy and Implementation Issues. New York. Klare, B., Burge, M., Klontz, J., Bruegge, R., & Jain, A. (2012). Face Recognition Performance: Role of Demographic Information. IEEE Transactions on Information Forensics and Security, 7(6). https://doi.org/10.1109/TIFS.2012.2214212 Knopper, S. (2018). Why Taylor Swift is using facial recognition at concerts. Retrieved March 10, 2019, from https://www.rollingstone.com/music/music-news/taylor-swift-facial-recognitionconcerts-768741/ Larson, C. (2018). China’s A.I. Imperative: The Country’s Massive Investments in Artifical Intelligence are Disrupting the Industry-and Strengthening Control of the Populace. Sciencemag, 359(6376). Lee, D. (2018). TSA lays out plans to use facial recognition for domestic flights. Retrieved March 10, 2019, from https://www.theverge.com/2018/10/15/17979688/tsa-precheck-facial-recognitionairport-cbp-biometric-exit Levin, S. (2018). US government to use facial recognition technology at Mexico border crossing. Retrieved March 6, 2018, from https://www.theguardian.com/technology/2018/jun/05/facialrecognition-us-mexico-border-crossing Lewinski, P., Fransen, M., & Tan, E. (2014). Predicting Advertising Effectiveness by Facial Expressions in Response to Amusing Persuasive Stimuli. Journal of Neuroscience Psychology and Economics, 7(1). https://doi.org/10.1037/npe0000012 Liao, S. (2017). Facebook uses selfies as login authentication for suspicious activity. Retrieved March 10, 2019, from https://www.theverge.com/2017/11/29/16716278/facebook-tests-selfieslogin-verification-face Lynch, J. (2018). Face Off: Law Enforcement Use of Face Recognition Technology. California. Patel, C. (2014). 3 companies promoting the use of facial recognition in payments. Retrieved March 10, 2019, from https://gomedici.com/3-companies-promoting-use-facial-recognitionpayments/ Quintas, P. (2018). The Truth and Benefit of Using Facial Recognition in Law Enforcement. Retrieved from https://medium.com/10-eight/the-truth-and-benefit-of-using-facial-recognition-in-lawenforcement-55c029080c23 Sharman, J. (2017). Metropolitan Police’s facial recognition technology 98% inaccurate, figure shows. Retrieved March 12, 2019, from https://www.independent.co.uk/news/uk/home-news/metpolice-facial-recognition-success-south-wales-trial-home-office-false-positive-a8345036.html Tolba, A., El-Harby, A., & El-Baz, A. (2005). Face Recognition: A Literature Review. International Journal Ofo Signal Processing, 2(2). van Dijk, J. A. G. M. (2019). General Conceptions on the Impact of Technology. Enschede, Netherlands. Whitener, M., & Aragon, R. (2019). How should we regulate facial-recognition technology? Retrieved March 12, 2019, from https://iapp.org/news/a/how-should-we-regulate-facial-recognitiontechnology/ Windows. (2015). What is Windows Hello? Retrieved March 10, 2019, from https://support.microsoft.com/en-my/help/17215/windows-10-what-is-hello Winston, A. (2013). Facial recognition, once a battlefield tool, lands in San Diego County. Retrieved March 11, 2019, from https://www.revealnews.org/article/facial-recognition-once-abattlefield-tool-lands-in-san-diego-county/ Xie, M. (2018, September). The Future of Biometric Facial Recognition. Retrieved March 10, 2019, from https://www.forbes.com/sites/forbestechcouncil/2018/09/19/the-future-of-biometricfacial-recognition/