Facial recognition could help prevent motorcycle theft

By ETX Daily Up | 13 April 2023


MUNICH: BMW Motorrad will soon offer facial recognition on upcoming models in its Boxer range, a world first. The concept behind the innovation is to fight against theft with a system that only allows the vehicle to be unlocked when the driver is recognised.

BMW Motorrad is believed to be the first motorcycle manufacturer in the world to introduce this type of technology, which it has named BMW iFace.

This facial recognition system uses a 3D technology directly integrated into the TFT display of the motorcycle.

Once the helmet is removed, the face is biometrically scanned in three dimensions. This three-dimensional image is then compared to pre-registered data stored in the system.

If the pilot is wearing a helmet, identification can then be done by scanning the iris and cornea of the eyes. In all cases, if the individual is recognized, the ignition, steering lock and other locking functions are released and the rider can start up the motorcycle.

The system has been honed to function regardless of the lighting conditions, day or night. According to the manufacturer, a special filter allows eyes to be scanned even through strongly tinted visors, as well as through types of glasses and contact lenses.

The system has additional security features; for instance if someone attempts to steal the motorcycle, BMW iFace then communicates with the eCall emergency call service.

In such cases, the data from the face or eye scans as well as the GPS position of the motorcycle will be automatically transmitted to the police authorities (initially in Germany).

BMW iFace is expected to be officially launched in the fall of 2023, exclusively on BMW Motorrad Boxer models.

While this may be a world first in motorcycles, such a system also exists in the automotive domain, on the Genesis GV60 and the Zeekr X.

This is the first solution for unlocking your car by facial recognition, directly implemented in the vehicle, to be rolled out on production models.

Keywords