Gait Forensics: Why the government and their cameras will get it right

On Podcast 17 (click here for a link to our iTunes account) we mentioned and shared a brief snippet from a recent Nova Now (video.pbs.org) entitled “How the Brain Works”. In this podcast we discuss astrophysicist extraordinaire Neil DeGrasse Tyson’s thoughts on how vision can fool us because of 2 basic types of neuroreceptors, one that detects motion while the other suppresses the background. This occurs when we are tracking an object in something called smooth pursuit.  In other words, the brain doesn’t pay attention to everything our eyes look at.  The brain just cannot process all of the visual information accurately. This is a handicap as a human. It is what can make us good at some aspects of vision and less good at others, the question is “Can this be taught and refined ?"  Great basketball players for example can see the the entire court and see plays developing. Wayne Gretzky was also reportedly great at this. This is what led him to say "I skate to where the puck is going to be, not where the puck currently is."  Watch any youth soccer  game and you will see just the opposite. You will see a hoard of little arms and legs all huddled around the soccer ball, no play is developing rather they are all in the moment competing to get the soccer ball.

Why are we talking about this from this perspective ?  Well there have been continued developments in the government’s gait recognition software, something we refer to as "Gait Forensics”.  Here are 2 links.

1. http://www.dnaindia.com/scitech/report_now-cameras-can-identify-thieves-from-their-walk_1743650

2. http://rt.com/news/identify-walk-system-britain-668/

Computers will likely have the advantage of gathering all of the info on a persons gait. It will not get caught up on the face or the clothes, there will be no discrimination. The software will likely capture head carriage, arm swing, stride length, step length, cadence, postural characteristics and so much more.  It will be more information than the human will likely be able to process because computers will not likely suppress any aspects of a persons gait like we mentioned in the opening paragraphs of this blog piece. To the computer, all  parameters will have equal and top priority, at least in gathering information.  Interpretation remains another matter. But we are most sure that in time that too will be ironed out.

Shawn and Ivo