What is eye tracking?

Eye tracking is a sensor technology that enables a computer or any other machine to determine what the user is looking at – the gaze point.

Home / Technology / What is eye tracking?

Placed on or within the machine interface and using the eye’s reflection of near infrared light beams, eye tracking technology calculates data about the user; detecting presence, attention, and focus as well as the position of a person’s eye and pupil size.

Eye tracking data provides unique insights into human behavior and conditions, digitalizing the way people interact with machines and devices, providing a platform for enterprises to develop innovative solutions and applications.

The technology in itself is quite straightforward. The magic, however, lies in being able to carry out eye tracking in a way that it can be used by everyone, across the world, under a variety of environments and conditions with computational efficiency, high precision, and within the limited space posed by modern devices and machines.

Eye tracking components typically include one or more high resolution cameras, near-infrared light sources, and algorithms that translate head and eye movements into real-time data streams, leveraging machine learning and advanced image processing.

The light sources create reflections — what we call glints — on the cornea of each eye

At a high frame rate, the cameras take images, recording the person’s eyes and the glints.

The algorithms use machine-learning techniques to identify the pupils and the glints.

Using advanced image processing and knowledge of human eye-anatomy the algorithms can calculate data points, such as pupil position, gaze vector of each eye, gaze point, and eye openness.

A straightforward technology

While simple, this concept is founded on many years of research and development both at Tobii and other research institutes. Building an eye tracker is simple. Building an eye tracker that works for every user around the world is not.

The eye tracker emits a near infrared (NIR) light beam

This light is reflected in the user’s eyes

The reflections are captured by the eye tracker’s cameras

Through filtering and triangulation, the eye tracker determines where the user is looking—the gaze point—and calculates eye movements data.

On the inside

The eye tracker consists of sensors—a camera and projectors—as well as image processing algorithms. The sensors, consisting of custom components and advanced optics designed for performance:

Advanced micro projectors beam Near-Infrared (NIR) light, creating a reflection on the user’s eyes
Sensors capture high-frame-rate images of the user’s eyes and light reflections
Image-processing algorithms the intelligence of the system, detect useful details in the user’s eyes and reflections, interpreting the image stream to determine the position of the user’s eyes and their gaze point on a screen.

The magic of eye tracking

A device or machine equipped with an eye tracker can determine user intention because what a person looks at is a good approximation of what they are thinking. In other words, eye tracking can provide insight into the human mind that can be used in a variety of applications:

To humanize technology by making it more intuitive
To increase computing and resource efficiency
To learn from experts, enabling training and skills transfer
To assess medical and human conditions and behaviors

Many amazing new applications are waiting to be developed. Can you imagine how your innovation could benefit from eye tracking?

Humanize technology

Understand user intention—devices and machines equipped with eye tracking are aware of user focus. This knowledge can be leveraged to create intuitive user experiences for any computing system. It is fundamental to the creation of humanized technology.

Create new user experiences and natural user interfaces—on its own or by combining eye tracking with other input mechanisms, such as, keyboards, mice, joysticks, touchpads, and voice commands.

The addition of eye tracking to machines makes interacting with them seem more natural and engaging than with conventional interfaces. Think about how you identify and select icons on your screen. If the computer automatically identifies the icon you want to select by using your gaze point, the need to grab and point with a mouse becomes redundant. Eye tracking data is a way of digitalizing the way we communicate with machines and automating manual tasks.

Resource efficiency

Armed with the knowledge of where a person is looking on a screen, computational resources can be focused on that spot. The human eye processes only small area with sharp focus—imagine a circle, with a diameter slightly under an inch, held at arm’s length.

Our brains approximate the rest. This fact can be used to reduce the need for high resolution graphics and rendering power to the area where the user is focused. By exploiting the knowledge about where a person is looking, and perhaps more importantly not looking, computing efficiency and data transport can be optimized.

Efficiency is not limited to calculations on the interface, it can be applied to other critical resources within your system. Consider, for example, a clinical device that uses the surgeon’s gaze point to guide the taking of X-Rays during surgery. Limiting the captive area of the X-Ray to the point where the surgeon is focused reduces exposure to both patients and clinical staff.

Training, skills transfer, and documentation

One of the best ways to acquire new skills is to learn from someone else, by watching what they do and trying to emulate that.

With eye tracking, training, skills transfer, and collaboration across organizations take on new scaling possibilities. Imagine, for example, a novice equipped with an eye tracking enabled headset or glasses. An expert or instructor can remotely guide them through whatever process needs to be completed, in real time. Such scenarios can be of use say in equipment repairs out in the field, or in a server hall. By recording the gaze pattern of the expert, more people can learn and follow, solving complex tasks in almost any domain.

Eye tracking can be used to advantage in scenarios that require people to rapidly scan objects during quality control and fault-checking processes. Such processes arise in most industries, but eye tracking enables them to be measured and digitalized, data that can be used to document and automate manual processes.

Condition assessment and wellbeing

Given that eye tracking provides data about a person’s presence, attention, focus, drowsiness, consciousness, and other mental states, it can be used to assess certain medical conditions.

Once current status has been established, the measurable and quantifiable metrics provided by eye tracking can be used for follow-up purposes, to, for example, assess the success rate of treatments. Best of all is that eye tracking data can be obtained in a non-intrusive manner, making it more comfortable for patients, and without the bias and errors that occur during manual observations.

Find out more about our products