At Seeing Machines, we set ourselves the long-term goal to build machines that can better understand the humans they serve.

As humans, we all share an innate ability to watch the faces and eyes of others, to help us better understand their minds.

Indeed, so powerful is this “facial” sense of understanding, that we even look at animals such as our pets and convince ourselves of their human-like personalities. Questions like: What is that person feeling? Do they understand me? Do they care? Do they look tired? These can all be answered without words, by us humans, sometimes with nothing more than a single glance.

At Seeing Machines, we set ourselves the long-term goal to build machines that can better understand the humans they serve.

This goal is to develop “machine empathy”. The reason we set ourselves this goal, is for the same reason human beings need empathy; to work well with others. Machine empathy brings machines closer to our human world and makes them more intelligent to our needs.

Seeing Machines

What we do

Seeing Machines specialises in computer vision algorithms, optics and processing technologies that enable new forms of intelligent machines. In particular, we are concerned with machines in ‘mission critical’ applications, where accidents are unacceptable and where safety can be enhanced through monitoring of the human operator, driver or pilot who is ultimately responsible.

Today, Seeing Machines’ FOVIO Driver Monitoring (FDM) processor employs computer vision algorithms that have been developed over more than 20 years of applied research.

The algorithms can robustly, accurately and in real-time, measure a driver’s visual attention to their environment, assess their degree of drowsiness, and ultimately detect if the driver has passed a threshold of risk. Mitigating the risk is then enacted through intelligent alerting of the driver as well as informing a vehicle’s wider control systems.

What sets us apart

DMS (Driver Monitoring System) solutions built on Seeing Machines’ FDM technology are automatic, unobtrusive, accurate, reliable and intelligently aware. There is no need for the driver to wear anything or modify their behaviour. The technology does not require cooperation, calibration or any prior knowledge of the driver.

Fundamentally, Seeing Machines technology does not just track faces and eyes, it solves real-world problems and saves lives.

Distraction

Distraction has many causes, the most common being phone use, but can also be due to vehicle controls, passengers, other vehicles, eating, or even day-dreaming. The potential causes are endless.

Whatever the cause, if the driver, pilot or vehicle operator is not observing the operating environment, such as the road, frequently enough, then the risk of an accident will greatly increase.

The face and eye tracking algorithms in Seeing Machines FDM processor feed a Driver Distraction Event (DDE) algorithm which must be robust to face masks, sunglasses, facial expression, behaviour, hats, wigs, jewellery and all the myriad of human behaviours which might indicate distraction.

Drowsiness

Drowsiness is characterised as a mental impairment due to lack of sleep. It cannot be overcome by just “waking up”, the only solution is rest. Drowsy driving has been a long-standing safety issue in nearly all transport markets. Unlike alcohol, which can be breath-tested, there has been no reliable way to measure the degree of drowsiness of a driver with any degree of certainty.

However, after over a decade of research and leveraging best-in-class underlying head and eye tracking algorithms, combined with the latest machine learning techniques applied to carefully gathered datasets, Seeing Machines is now able to provide a Driver Drowsiness Level (DDL) signal that offers an actionable assessment of risk of the driver losing control of the vehicle in the near future.

Microsleep and Incapacitation Detection

Everyone gets tired a little differently. Seeing Machines technology watches out for signs of the driver passing in and out of consciousness; so called “microsleep” events.
Some driver’s never exhibit signs of drowsiness or even microsleeps and can just suddenly pass out. This can also occur due to medical reasons, such as low blood sugar. Seeing Machines technology is able to detect a driver that is unconscious.

Vigilance and Engagement

Conditionally autonomous vehicles and advanced safety functions such as collision avoidance systems require a reliable measurement of a driver’s level of “engagement” to the driving task (or the road scene). We call this Driver Engagement Level (DEL). This signal is designed to allow vehicle systems to intelligently and safely hand over control, or take control back, from the driver.

Unrivalled DMS solution performance

The FDM processor sits at the heart of modern camera-based DMS. It offers unparalleled measurement quality of all the levels of driver impairment and engagement, with algorithms that have been calibrated against diverse population studies, examining the ability of driver’s to safely control vehicles in complex road environments when in known mental states.

The FDM inter-operates with Seeing Machines specialised and highly tuned optical reference designs. These are infrared camera solutions that employ dynamic sensor and illumination control by the FDM monitoring algorithms. This tight coupling between algorithm and optics enables the DMS system to track a person’s eyelids and pupils through sunglasses under every imaginable real-world driving condition.