“Anything new is at first anything bad and then everything necessary.” This is the universal process of admission of change, anywhere. Any innovation that comes to the humankind for scrutiny, has to first go through an extensive process of analysis, criticism and then finally acceptance. Then how can we not jump down the same tunnel and even more enthusiastically, when such a change is induced in instruments that are inseparable parts of our beings?
The Facial Recognition feature now streamlining into all of your mobile devices, adapted by the IOS and Android alike is one such innovation that must be put under the microscope.
Passwords, patterns, fingerprints and now this facial recognition, the timeline of security measures put forward by your mobile devices have leaped through several revolutionary changes in the past decades. There was a time when just to open your phone, you had to go through the cumbersome process of typing in a combination of alphabets, numbers, and symbols and god forbid if you forgot! Of course, this was never meant to be a long-term solution, even the patterns which made this job considerably easier never passed the bar in security or feasibility. The biometric methods, on the other hand, came riding as the knight in shining armor rescuing the users from the mundane waste of time and effort. With these you no longer had to go through the tiresome process of typing, erasing and typing again and you could just as easily have let go of trying to remember new passwords or worry about people guessing your passwords.
But just like all stories, its not all sunshine and daisies. The facial recognition feature has its own share of issues, of which you may hear the users constantly complain about. The primary and most common issue is that of the camera angle. While setting the face ID, most devices ask for the scanning to be done at a various angle so as to avoid any trouble in recognition later making the setting of the ID a cumbersome process in itself, but that’s not all. The face ID may not match the said persons face even when there is some sort of obstruction present in the picture, like hair, glasses, etc. So just imagine having to hold up your hair or removing your glasses each time you have to open your phone! It sounds exhausting, doesn’t it?
Another featured flaw of this is the storage space, it requires a lot more space to save the data regarding the facial recognition than it does in a simple password or a pattern and in an era where there is so much data available to store and dispose of, every “bit” counts. The image stored and scanned each time is that of a high resolution, the process used in extraction, comparison, and matching is in itself a heavy one, thus not only does it occupy a lot of space in your device’s memory but just carrying it out each time extracts a lot of power from your device, not to mention the time delay that happens in the carrying out the overly complicated process.
The next fatal flaw, I believe in this feature is somewhat temporary- the requirement of high-quality cameras. While I phone X users, may not face this problem, but those using a cheaper, less advanced phone will definitely be sharing these woes. The facial Recognition feature is based upon a solid method of scanning, extracting, comparing and matching. It selects about 80 nodal points in the face, including the length of the jawline, etc. and stores it in its memory. But, when the camera in itself is incompetent, or the image captured by it is comparatively blurry or unclear, then all of these processes take a hit. The general efficiency of the process declines with each step ebbing a certain percentage out of it. So Facial IDs can only be afforded by users of expensive devices. The inaccessibility of this feature, though in my opinion is a temporary problem, still remains one of its biggest flaws.
So all in all, I believe, that while in theory, the facial recognition feature is quite a messiah, saving us from wastage of time, going through complicated processes, etc. and adding a greater sense of security and accuracy in accessing our devices. In reality, there is a lot to be done. The feature has to travel a long and arduous path of development, the challenges in front of it are not undefeatable, and once done, it would become a feature uncontested by any other, standing tall among its peers.