How this technology is used by states and businesses, whether it is possible to trick a camera with a face identification system.
For the state, facial recognition is an important part of the security system and an impressive budget item. For journalists, it is either a panacea or a tool of a global conspiracy. For business – a tool or product. Whichever side you take, the basic questions still remain. Users usually search for answers to them on the Internet (on average, 28,704 requests on the subject of face recognition per month), but they do not always find them. Fixing the situation.
What is face recognition?
Users are more likely to encounter face recognition in their smartphones, where biometric identification is used to unlock the device and only the owner could access the data. In the recognition process, a 3D camera must be involved, so that it is impossible to deceive the gadget with a photo.
There is also the identification of faces in real-time and in real conditions: in this case, it is inextricably linked to video surveillance systems, where faces are literally “snatched” from the video stream captured by cameras.
Imagine a high-quality modern video surveillance camera placed just above the average human height in a well-lit place. In front of it, every day passes about the same number of approximately the same people. They don’t move very fast.
The captured video can be stored in a cloud archive. An analytical module is connected to the camera: a complex combination of algorithms (artificial intelligence, neural networks, that’s all) plus a user interface. The module “snatches” faces from the video stream, determines gender and age, and enters the data into the database.
Gradually, there are more images. The system remembers all recognized faces automatically and enters them in the archive, and the user with permission specifies additional data: name, position, status, and other marks (“VIP guest” or “thief”). You can upload a photo of the desired person, and the module will find all the detections of this person in the archive.
As soon as the person with the mark passes in front of the camera again, the system records this as an important event and sends a push notification to users.
Detection in the context of face recognition is a situation where the algorithm basically realized that it was a face, and not an apple or a mermaid from a Starbucks mug. He first needs the computing power to do this, and only then can he match the face with the base or remember it.
Face recognition doesn’t always work correctly
Sometimes detection can surprise you. If you’ve read the last few paragraphs to the end, congratulations, now you know how to face recognition works in an ideal situation. The description is suitable for any system: from those used in the underground to solutions for small businesses.
The main thing to understand is that it is difficult to create an ideal situation in real life, especially if we are talking about the whole city, and not an office or a store. For example, there are a lot of people in the underground, moving fast. You need a lot of cameras, they cost money, and they should be placed by competent specialists.
Is it possible to cheat the face recognition algorithm. Despite the mistakes that occur, the accuracy of machine recognition often exceeds that with which people determine faces. China will soon build a giant facial recognition database to identify any citizen within seconds, a system that can find a specific person among 1.3 billion other residents in 3 seconds with 90% accuracy.
And yet it is difficult to answer this question unambiguously because there is no single ideal algorithm for facial recognition. Large glasses, pasted beard, cap, high-speed movement, makeup (for example, painted on the face of the grid “Black Swan”, seals, circles, and sticks. All this can confuse the algorithm. Especially in the aggregate, because for recognition, it is enough to deceive the recognition systems of about 70% of the open face. Now imagine that you need to use the above tricks in a real city. That doesn’t sound so easy, does it?
How face recognition is programmed
Face recognition (as well as other related operations) is a fairly common task. Therefore, many companies provide ready-made services in the form of cloud APIs (software intermediaries between applications) for high-quality solutions to these problems. In addition to IT giants like Microsoft and Google, specialized companies, are also engaged in face recognition. Their products are rapidly evolving and provide even more interesting features, such as identifying faces and silhouettes in a crowd.
It is much more difficult to train a neural network from scratch. You need a large and high-quality set of source data, that is, tens and hundreds of thousands (and even more is better!) photos of people. In addition, you will need significant computing resources and knowledge in the field of AI and machine learning. Large companies have all these tools, so they solve the problem much better.
There is also an intermediate solution – to use an already trained neural network, such as OpenFace. This option is likely to work a little worse than a ready-made cloud service, but it will allow you to have full control over the system. This will require a certain level of understanding of the work of neural networks and neural network frameworks, and, most likely, some knowledge of the Python language, which has gained popularity as the main programming language among Data Science specialists.
Indeed, it is convenient to conduct various experiments, visualize data and perform efficient matrix calculations thanks to the excellent NumPy package. This is not the best language for industrial development, since it does not contain effective tools for creating large, secure software systems, but there are no alternatives to it in the field of deep neural network training yet.
How does face recognition work in business
The demand for face recognition in fintech, retail, and other businesses is directly related to the increased availability of the technology. The mechanics are simple: all enterprises and all organizations have video surveillance cameras, which are used as tools for data collection and subsequent analytics. In the world, surveillance systems shoot terabytes of video in Full HD format per month, which means that a lot of information is accumulated for processing.
The necessary software for data analysis can be “flashed” to the device by the manufacturer. Cameras with video analytics “on board” are usually quite expensive.
An alternative option is an analytics in the cloud, that is, a remote data center that connects to any inexpensive camera. This is an order of magnitude cheaper, plus it gives you flexibility – you can adapt solutions to a specific business.
The popularity of facial recognition technology in various fields of activity is increasing.
Another thing is that announcing, testing, piloting, and buying solutions do not mean implementing them.
With retail trade, everything is more transparent. There are three problems here that face recognition solves.
First, theft. Scammers operate in stores, and often the same people in the same network. Face recognition allows you to identify “drifting thieves” and other people who have previously violated the order. As soon as the violator once entered into the database enters the store, the security guard will receive a notification in the messenger or in another convenient way.
Secondly, the difficulty of working with regular customers. There is simply not enough data on purchases and birthdays to personalize offers for VIP customers and fans of the brand. Face recognition can be integrated with CRM-that is software in which managers enter all the information on all the transactions of the organization. In the case of thieves and VIPs, face recognition works in much the same way: the face is blacklisted or whitelisted, and when it reappears, the system will signal the person with access. Gender and age are determined automatically, and additional information will be added by the responsible employee.
Third, the identification of individuals in retail is used for targeted advertising. For example, in some stores, X5 Retail Group installed X5 will enable computer vision cameras to recognize the facial expressions and age of customers. Analyzing this data, the system displays on the monitor screen the trading floor products that a person may like. More live illustration – Lolli case & Pops, a large candy store in the United States. The face recognition system detects the future in-store loyalty program will be fed by facial recognition of regular customers and sends notifications to their smartphones with products that they may like (taking into account individual preferences and even food allergies).
Another vivid example of the use of technology in retail stores without sellers and cash registers. For example, Alibaba Tao CafeAmazon Go vs Alibaba Tao Cafe: Staffless Shop Showdown is a cafe and self-service store located in Hangzhou. It sells drinks, snacks, food, toys, backpacks, and the like. Tao Cafe is only open to users of the Taobao website.
Face recognition in the field of trade
When buying drinks, the face recognition-enabled camera system automatically identifies the customer, contacts their online store account, and processes the payment. Customers exit through a room equipped with several sensors that identify both the customer and the goods. The scan works even if the person has put the purchase in a pocket or bag.