April 2017 Digital Edition
March 2017 Digital Edition
Feb. 2017 Digital Edition
January 2017 Digital Edition
Nov/Dec 2016 Digital Edition
Oct 2016 Digital Edition
OPINION / Visual modeling of faces for security applications
By Jennifer Tidy and Parham Aarabi
Virtual aging of photos is often used by law enforcement for finding and capturing fugitives, abducted or lost individuals, or even terrorists. However, on some occasions the photos are either unrealistic or inaccurate. Furthermore, these photos are often generated by combining multiple stock photos (i.e. hair from stock photo 1, eyes from stock photo 2, etc.). While such a feature cut-and-paste method may be easy to do, it does not often produce the most realistic results.
In the past few decades, there has been an exponential growth in the anti-aging and cosmetic surgery space, with billions being spent by consumers to reduce or remove the signs of aging on their face and body. Plastic surgeons and pharmaceutical scientists have been exploring new ways to reduce wrinkles, to redraw the facial contours, and to adjust the tightness of the skin and restore the firmness of the facial muscles. Companies have begun to explore ways to show anti-aging and cosmetic surgery effects, using computer modeling on user photos, to aid both interested consumers and cosmetic surgeons. These programs are often based on physical modeling of the face, and can result in highly accurate renditions of a wrinkle-reduced face.
But what if instead of reversing the aging clock, we wanted to move it forward (while using the same physical modeling)? Perhaps there will never be a need for a pro-aging cream or a cosmetic surgery that makes us look older, but the technology used for visualizing the effects of anti-aging can be very readily applied to showing an older you.
Instead of simulating the addition of color to your hair, this technology could show grey tones added to your hair. Instead of virtually lifting the facial muscles, this technology could visualize drooping jowls and loss of volume that results from aging. And finally, instead of detecting wrinkles and removing them, these wrinkles could be strengthened and deepened. An anti-aging virtual makeover, turned inside out, would result in an extremely realistic and accurate aging makeover.
Which brings us to our central point, why have recent law enforcement aging applications not been as accurate or as realistic as one would hope? The first part of the answer is that, for law enforcement, just aging a photo is not enough. They need to think about disguises and changes to the face which would need to be shown on the image. The second reason, however, is that physics-based modeling of facial aging has not been widely used. Once such aging and facial modification programs become more available and easy to use, they will enable law enforcement to provide more accurately aged photos, which hopefully will increase their chances of capturing fugitives, finding missing persons, or even entertaining the curious who want to know what they will look like as they age.
Parham Aarabi is CEO of ModiFace.com and Jennifer Tidy is partnerships manager.