Quote Originally Posted by Thailandnotes View Post
There was a fascinating article on surveillance cameras in London and law enforcement personal who have prosopagnosia vs other personal who are almost savants.

There's a 20 minutes test at the end of the article. 2 weeks ago in the New Yorker.

"The average human can recognize a family member from behind. No computer will ever be able to do that.”

http://www.newyorker.com/magazine/20...r-police-force
From what I've seen/heard I would tend to disagree.

A computer algorithm unable to positively match a human photographed from behind as a still image I would agree.

But live video is a lot easier to match suspects if there is comparison video to analyse and catalog characteristics like movement/gait/arm swing.

The way people move can be unique-ish