Software identifies dodgy crime scene gawpers

It’s true, apparently, that criminals – especially arsonists – often return to the scene of their crime, as do IED bomb makers in the Middle East.

And a team of computer vision experts at Notre Dame University say they’ve developed a way to help identify suspicious individuals at crime scenes.

The researchers’ ‘Questionable Observer Detector’ (QuOD) is designed to identify people who repeatedly appear in video taken of bystanders at crime scenes.

It’s not something that most facial recognition systems can cope with, as it needs to work without an existing database. It also needs to be able to deal with poor-quality video shot by amateurs.

QuOD works by creating ‘face tracks’ for all individuals appearing in a video, and repeating the process for all available video clips. The face tracks are compared, and if any faces from different video clips look similar enough to match each other, a cluster is created.

An individual is considered suspicious if law enforcement considers that he or she appears just that bit to often for comfort.

There are still problems to overcome – lighting and resolution for such bystander videos are often very poor indeed, and the sheer number of videos to be processed can often cause difficulty. But the team say they’re confident of overcoming these challenges.

Far more important for many, though, will be the issue of civil liberties. The researchers say they believe people will approve, given that the system identifies people who were actually present at crime scenes, rather than simply suspected of being there.

They may be being a little optimistic.