By adding an audio signal to such a system, Sharma has demonstrated that its accuracy can be increased significantly. Some day, perhaps, an airline pilot may just say "hi" to a camera and be cleared to pass through the security gate without even so much as a pause. The system will recognize precisely who she or he is.
Sharma sees all sorts of other applications, such as marketing research. It could tell whether more men or more women climbed behind the wheel of a fancy sports car on a showroom floor, or which gender favored a particular brand of coffee.
These advances are always a bit troubling to some folks who see computers as an easier way for others to invade our privacy.
It probably won't be long before a system like Sharma's will gaze into our eyes and tell the sales person whether we can make the payments on that fancy car. Or it may even tell our significant other that we fell off our diet and swiped a chocolate eclair while we were supposed to be picking up the broccoli for dinner.
But there's lots to be said for systems that can tell us, for certain, that the person sitting in the pilot's seat deserves to be there.
Lee Dye’s column appears weekly on ABCNEWS.com. A former science writer for the Los Angeles Times, he now lives in Juneau, Alaska.