Face recognition systems started in 1960, Bledsoe developed a method to classify photos of faces. The system could record the coordinated location of various facial features. Then a matrix would be inserted into the database, and the system would give a closely resembling photo. Why is this relevant? Because this is where we started to consider Face recognition as a viable biometric. In 1991, Turk & Putland discovered how to detect faces within images. In 2010, Facebook implemented Face recognition functionality and rolled out tag suggestions. And in 2017 iPhone X released Face recognition as a feature for device security, subsequently making it the gold standard for safety.
Ever since digital cameras came around, we all took photos, tagged each other, and posted them on the internet. The unlimited number of pictures and geotagging built the mind of the Face recognition algorithm. The algorithm locates the face on an image, maps facial features to correct head rotations, and takes multiple measurements that define the face in raw pixel data.
Face recognition is solving several issues that have been long pending. One such smart use of Face recognition was developed by Megvii. Megvii is a Chinese technology company that designs image recognition and deep-learning software for the public sector.
Megvii AI-powered surveillance system scans streams in real-time, helping detect danger. SenseTime, a leading global artificial intelligence (AI) software company, announced that its next-generation Artificial Intelligence Data Center (AIDC) had commenced operation. SenseTime’s SenseFoundry software platform, powered by AI models, transcribes raw and real-time city visual data into insights, alerts, and actions which means that any unusual situation can be tracked and diffused in real-time. Still, it also tracks the behavior of crowds in an area. For example, how many people are inside a building at a given time frame? This information comes with a lot of data that, if not protected, can be damning to us, everyone.
The scary part of this story is the data on people, patterns, where they are going, etc. This fundamentally changes the relationship between the government and the citizens. In India, on January 20, a lawsuit was filed in the southern state of Telangana by a social activist. After being forced to remove his mask by the authorities for a picture in a heavy camera-surveillance-equipped city. Telangana state has more than 600,000 cameras, most of which are in the capital Hyderabad. The lawsuit is a test case as facial recognition systems are deployed nationwide. The activist S.Q Maqsood’s said, “It is also about my right to privacy, and my right to know why my photograph was taken, what it will be used for, who can access it, and how it’s protected. Everyone has a right to know this information,”.
This quote aptly answers the question of our anonymity. It asks the right questions about privacy, infringement, and the right to obscurity. Obscurity has always been an essential component of privacy, meaning that personal information isn’t readily available to anyone. It doesn’t mean that information is wiped out or even locked up; instead, it means that some combination of factors makes certain types of information relatively hard to find.
The future of Face recognition today looks bright and vulnerable. The facial recognition market is expected to increase by $9.6 billion by 2022. Advancements in the next generation of tech with software like Coresight appear bright and promising. Coresight facial recognition technology works with face masks to allow contactless biometric entry for medical employees and others. And lastly, they are vulnerable because of the regulations and chances of being strictly regulated or completely banned.
