So I got a new laptop over break and I got this cool software, Veriface, already installed on my computer.  What it does is let’s the user use his or her face as their password (for log in, or even certain applications or documents).  I asked myself, how does this work?  It turns out there are many ways, but one way in which facial recognition can be implemented with good accuracy is using Interpolated Bezier Curve Based Representation .  Basically it uses some complex algorithm to compute Hausdroff distances (discussed further in the paper), which in turn computes a more accurate function representation of the face.  The article even says “The recognition rate is 100% on normal faces and also on faces with some simple expressions,” which is staggering above previous implementations of facial recognition years ago.  This software is becoming ever more popular, not just on notebooks for personal security, but even private companies such as clubs, bars, and banks.  This software has even been emerging in ATM’s specifically to heighten security.  Now even if someone stole your card and knew your pin, they still couldn’t withdraw money from your account!  An article even goes as far to say that the government used this software during SuperBowl XXXV to check for would be felons or terrorists (http://archives.cnn.com/2001/TECH/ptech/07/17/face.time.idg/).  As this software is not 100% accurate all the time, it is vastly superior to things like iris scanning which even though are extremely accurate, actually need the user to be very close to the scanner.  This facial recognition software allows a clever way to make matches at a distance, increasing security as well as comfort.  As these methods become more accurate (such as the Bezier Curve Interpolation), facial recognition security methods will become extremely practical and become a much more stable security method rather than passwords and keycards.  I could even see in the future that a webcam might be needed to confirm an online credit card purchase so that they can match your face against their database (or the credit card’s database of you).  This could drastically reduce identity theft and online credit card fraud.  An example of how the facial features are computed.(primidi.com)