Eye tracking is the process of measuring eye movements in order to determine where a person is looking, what they are looking at, and for how long their gaze is in a particular spot. The technology is capable of converting eye movements into a data stream that contains information such as pupil position, the gaze vector for each eye, and gaze point. The insights developed from the translation of eye movements can further be used in a range of applications or as an input modality.
Often, eye tracking systems use invisible, near-infrared light and high definition cameras to project light onto the eye and record the direction it is reflected off the cornea. Algorithms are then used to calculate the position of the eye and where it is focused. This can make it possible to measure and study visual behavior and fine eye movements.
The brain processes images through light-sensing cells in the retina. These cells, rod cells, and cone cells detect light through the pupil and send visual data to the brain. While there are fewer cone cells than rod cells; the former permits a person to see in high resolution. Cone cells are in the fovea, the center of the visual field, and is the region with the highest visual acuity in the eye, but it is small. Outside the area of the foveal vision is the peripheral vision, where clarity decreases. This is why the eye has to move in order to process detailed visual information of interest. These movements are what is tracked by eye tracking technology, among other observations.
The eye executes a number of movements, including vergence or torsion, but the most pertinent ones measured through eye tracking are fixations, saccades, and smooth pursuit.
- Fixations occur when the eye stops to collect visual data. Although the duration of one is highly variable, the longer a fixation is, the more visual information is processed.
- Saccades are fast jumps the eye performs between fixations in a static environment. The eye moves from one object of interest to another with the goal of acquiring new visual data in high resolution.
- Smooth pursuit is the eye movement that takes place when looking at an object in motion and following it. As visual intake is possible during smooth pursuit, the movement is relevant for eye tracking.
Video-based gaze tracking also allows measurement and analysis of pupil size. To obtain accurate pupil tracking, however, the environment needs to be rigorously controlled. In adequate test conditions, pupil dilation can be observed and monitored as a result of the following:
- Strong emotional stimuli
- Acute attention
- Illumination
- Working memory load
Perhaps of most importance, eye tracking facilitates the study of visual attention. Due to the limited nature of image processing sources, the brain selectively chooses relevant visual information based on:
- interest—the conscious or unconscious decision to look at an object, and
- environment—elements detected through peripheral vision.
In academia, the examination of visual attention leads to understanding of attention mechanisms in general. It sheds light on the cognitive processing that takes place during the execution of a given task. This same measurement makes it of importance and of value to the realm of marketing and advertising.
While the general principles of of eye tracking devices tend to be the same, there are several types of tracking devices. Which device is used depends on the nature of the research being conducted, with some being better suited than others to a specific research objective.
Types of eye tracking device
The category of embedded or integrated eye tracking devices is a growing category with increased use cases, some of which tend to draw the most interest from users. This has included devices aimed in surgery systems or related medical devices, autofocus camera systems based on gaze position within a camera's viewfinder, devices embedded into vehicle dashboards, and those integrated into virtual and augmented reality devices.
For research purposes, systems can be integrated into virtual reality and augmented reality devices to provide a control scheme and reduce variables into a study. This technology can also provide intuitive control method to menus within VR and AR technologies to eliminate the need of controllers or a mouse or keyboard.
Another possible use of eye tracking in virtual reality is foveated rendering. The human visual system is, as noted above, capable of high acuity at the fovea, and the peripheral vision has lower acuity and is incapable of seeing the same level of detail. This means graphical rendering in VR could focus on saving processing power by only rendering high quality graphics at the point of gaze, and lower quality in the periphery. However, this also requires embedded tracking systems fast enough to sample the movements and fast enough real-time data transmission to react to fast eye movements.
The human eye is constantly used for different tasks, and almost anything that involves a visual component can become a subject of eye tracking and the data collected by eye tracking devices can be leveraged to gain insights and understand human behavior. The following are use cases of the technology in this field.
