It’s a movie trope as old as cinema. A heated chase scene ends as the bad guy seamlessly blends into a crowd, pulling up his collar, safe in a swarm of people. But thanks to researchers at the Spanish National Research Council (CSIC), this scene may have a different ending; the pursuer pulls out a tablet and uses newly-developed software called idTracker to pinpoint the assailant’s exact location in the crowd. From a cinematic perspective it’s a slightly less exciting conclusion, but technologically it’s groundbreaking- there may no longer be safety in numbers.
Typically, computers have not been able to identify individuals (of any species) when they group together for more than a few seconds. But the CSIC team have developed new algorithms that may be able to change this. The abstract for the paper on idTracker in Nature Methods defines the project as “a multitracking algorithm that extracts a characteristic fingerprint from each animal in a video recording of a group. It then uses these fingerprints to identify every individual throughout the video. Tracking by identification prevents propagation of errors, and the correct identities can be maintained indefinitely”.
The current application of the software is mapping the behaviour of different animal species in groups. ‘Group mentality’ across the species is a widely-accepted concept, but the intricacies of group dynamics in different animals has been difficult to track and characterise up until this point. The CSIC research group have already used the software to track the behaviours of groups of fish, flies, ants and mice.
Alfonso Pérez Escudero, a CSIC researcher during the preparation of this study, stated “in the short term, this will be used in science, but in the longer term, the method we have developed can be applied to recognize people in large crowds, vehicles or parts in a factory, for instance”. Bad guys can rest safe among throngs of people, but only for now.