Researchers at the University of Washington are developing new camera technology based in smart cars that could eventually be used to track pedestrian and vehicle traffic on an individual level.
The project makes use of a fleet of smart cars equipped with the cameras. Each individual camera is part of a single network, and the images collected are all run through an advanced computer algorithm. The algorithm learns to identify people by comparing different camera angles, focusing on skin color, texture and movement. Using this information, the system can track an individual from one camera to the next.
As the technology develops, the research team hopes to coalesce the data into map form. It could also prove helpful to police, who could use the camera network to track suspects fleeing on foot or in a vehicle.
Researchers note that the camera network wouldn’t have to be limited to cars either, envisioning future tracking done via drone.