Semalt Expert: Current Trends In Eye Tracking Technology

When reflecting upon the early history of eye tracking and the milestones created, it has grown into a multifaceted industry that touches many fields. To further understanding what eye tracking is, all about it must first be defined.

Julia Vashneva, Customer Success Manager of Semalt Digital Services, gives insight into the latest trends in eye-tracking technologies drawing on expertise and extensive experience of cooperation with the CoolTool, market research company.

Eye tracking can be defined as a collection of ocular data. The data is gathered through non-intrusive devices such as a remote or head-mounted eye tracker. The non-intrusive device utilizes a camera as well as an infrared light source. The light source is directed towards the eye as the camera tracks and records ocular features (e.g., pupil diameter, blink frequency, rotational movements). The aggregated data is compiled by eye-tracking analysis software and results analyzed.

Eye tracking software is currently being utilized in consumer marketing and packaging research, academic, medical and behavioral psychological research, as well as human factors engineering and gaming simulation research and application. As researchers find new avenues of application for eye tracking, the field expands dramatically. Even the military has utilized eye tracking helmets for combat pilots in their sophisticated head-up displays (HUDs) for tactical advantage.

When discussing eye tracking, it is important to understand the four basic types of eye movements which is the basis of eye tracking technology they include:

  • Saccades. They are considered as very rapid movements of the eye which change abruptly at the point of fixation. They can be small movements during reading or large movements when gazing around an entire room. Generally, they are reflexive in nature, but can also be elicited voluntarily. Most common saccadic eye movements occur during an individual's sleep patterns.
  • Smooth Pursuit Movements. Much slower than other forms of eye movement, the Smooth Pursuit Movement is a voluntary stimuli movement of the fovea (a small section of the retina where visual acuity is the highest). The individual has the voluntary ability to either track or ignore the stimuli. There are those trained individuals who are able to make a smooth pursuit movement without the presence of a stimuli (moving target).
  • Vergence Movements. Unlike other types of eye movements, whereas both eyes move in the same direction, vergence movements are disjunctive. The lines of sight of each eye can see an object that is either close up or in a distance. The pupillary constriction thus increases the field of perception and sharpens the image of the fovea (retina).
  • Vestibulo-ocular Movements. Compensating for head movement, the vestibulo-ocular eye movements help to stabilize visual images of external impulses. The retinal surface compensates image "slipping" by detecting transient changes (head movement) thus adjusting through corrective eye movements. The compensation adjustment is rather slow, due to head movements as well as sensory response

Who Applies Eye Tracking

Every day, eye tracking applications are used in creative new ways. When one thinks about how much an individual uses eyes on a daily basis, it is no wonder that information it gathers is useful for almost every industry. One such an industry is market research. The information gathered from eye tracking research has been utilized in packaging design as well as advertisements.

Professional sports have been using eye tracking technology as an edge to improving an athlete's performance by combining biomechanical, physiological and behavioral data to provide a measurement tool for individual improvement. Behavioral and neuroscience professionals have been utilizing eye tracking technology in their assessing reading skills, dyslexia, motor functioning, as well as auditory and image processing. By the use of screen-based visual stimulation and head motion free tracking devices, researchers have found new ways to measure and assess and deficit motor functions as well as utilizing tactile approaches to correct individual deficiencies.

A new area of eye tracking is a gaze-based research. Researchers have been developing tools which may eventually change the interaction between human and computer. The research is focusing on using the eyes to control a computer's cursor. Instead of using a mouse, joy stick or writing tablet, the user will control computer actions via eye movement and blinking.

Exactly how eye tracking works for the common layperson can be explained in not-so-simple terms. Eye tracking since its beginnings has been used to measure and assess an individual's visual attention. There are several ways to measure such an attention, but the most common is the non-intrusive method.

Non-Intrusive Remote Eye Tracking

Non-Intrusive remote eye tracking's basic concept is the pupil center corneal reflection (PCCR). This concept utilizes a remote light source to illuminate the eye, which intern causes reflections and allows a camera recording system to capture imaging data. The image captured on the recording device (camera) identifies various reflections between pupil reflections and the cornea. Once information is captured an angled vector calculation table is used to form specific geometrical reflective features which forms the basis of gaze direction.

Depending on the eye patterning software utilized, the gaze direction calculation allows for various image-processing algorithms as well as 3D models of the eye in relation to the point of gaze. In rather simple terms, the process records eye pattern movements, assesses basic eye pattern principles, focuses on point of gaze parameters, and forms a compilation of data which when analyzed gives the end user specific interpretive data and results. The results can measure anything form a worker's attention span, to fixation on a specific object. To explain further, an example of the process is as follows:

  • The eye tracker researcher wishes to identify specific images on a computer (e.g., a marketable product). There are identical images of the product on a computer screen with different variations (color, size, etc.) The researcher wishes to identify which of the images the computer user focuses on (which in turn will be able to identify which product image is aesthetically pleasing to the user). The eye tracker system process includes a recording system, projector and a series of complex algorithms. Initially, the projector creates an infrared light image of the eyes. Next, the camera recording system takes multiple high-frame-rate images of the user's eye patterning (e.g., which of the images on the screen the user focuses on).
  • The imaging processing algorithms (specifically, eye tracking software specific to the image tracking) identify specific details of the user's eye and reflective patterns. Lastly, based upon the gathered imaging details the software calculates the eyes' positioning and gaze points. This data is then interpreted as to which of the images on the computer screen the user has focused their attention on (thus the researcher can identify which of the marketable products a specific end user group has focused upon).
  • As the field of eye tracking technology expands to various areas, questions arrive as to its effectiveness as well as concerns in obtaining accurate data. On such concern is whether blinking affects eye tracking results. As blinking of one's eyes is an involuntary action, it generally cannot be controlled. With regards to eye tracking, blinking appears to block the eyelid and cornea when the eyelid closes. As cornea reflection data is necessary in eye tracking data compilation, during final results analysis the missing data points are incorporated by either manual extraction or automated extraction by the eye tracking software. Although the data extraction may not be 100% it is close enough to draw accurate data conclusions.

Still another question arises as to whether head movement affect eye tracking results? Head movement during an eye tracking session is one factor which can skew results as it can have a definitive impact on the accuracy of gaze data. In solving the problem of head movement, multiple optical sensors and data gathering cameras are centrally located within the user's range of vision in order to create a type of “stereo data processing” which increases the horizontal range of vision. Once again, much like blinking, the eye tracking software will generate a schema which will take into consideration vestibulo-ocular (head) movements and adjust final interpretation data accordingly. What cannot be adjusted in the final results is if the eye tracker user turns their head to the side, institutes a fast and rapid head movement or is out of the range of the optical sensors and cameras.

When discussing eye tracker technology one question that always seems to arise is the accuracy and precision of eye tracking equipment (optical sensors, cameras, etc.) as well as data interpretation (software) analysis. These concerns have been addressed by various eye tracking companies who have developed measurement standards for ensuring the best accuracy possible. The Monocular Accuracy standard singularly tests each eye in order to ascertain the performance accuracy of individual eyes. As opposed to Binocular Accuracy tests which simultaneously tests both eyes (often considered the norm ratio).

The Future of Eye Tracking

Description Motive Accuracy considered the ideal testing condition is used as a baseline to measure all other accuracy results. Accuracy at extreme gaze angles tests the gaze angles affect which is the major component of eye tracking. Varying illumination accuracy is a test which utilizes four illumination criteria including foreground and background inversion (interference). Varying head positions accuracy which takes into account head movements during the eye tracking process. Each of these individual test standards in conjunction with equipment calibration are conducted in order to maintain the most accurate eye tracking results.

There are still those critics who have concerns that eye tracking is just a novelty technology. The critics have raised fears that the technology can be misused and/or the results of testing misinterpreted. As the technology has become more and more sophisticated and the results more defined, the critics seem to be few and far between. Eye tracking technology has been used to detect brain injuries and concussions, in education for those with reading disabilities and in most every field imaginable.

The future of eye tracking technology appears to be already here. Cellular phone eye tracking technology has already begun. The Samsung Corporation has introduced a line of mobile phones which will pause a front-facing video when the user looks away. Samsung and other mobile phone companies are introducing biometric iris recognition phones which will unlock upon the user gazing into the cell phone screen.

The automotive industry has been developing eye tracking attention monitoring systems which can detect when a driver is getting drowsy thus minimizing accidents. Accordingly, future automobiles may be equipped with eye pattern monitoring devices which can make adjustments in driving patterns by viewing the driver and the surrounding environment (rain-soaked pavement, snow conditions, etc.). Eye tracking technology has been introduced to the world of virtual reality gaming. VR systems use this technology to animate objects by the movements of the eyes. Within the medical field, the strides developed for eye-tracking will allow to detect the initial signs of strokes or seizures.

There many challenges that stand on the way of eye-tracking technology as it moves into the everyday mainstream of life. Eye tracking recognition software still has to differ itself from the simple eye movement (distraction vs intent). Other considerations to be addressed include vision variances, eye strain and distraction instances. As eye tracking monitoring equipment utilizes infrared beam exposures to the eyes, there has not been enough conclusive studies on the damaging effects of the prolonged or repetitive exposures.

For example, we in Semalt Digital Services are working closely with the CoolTool market research company, that provides a complete eye tracking technology along with other customer behavior tracking techniques for the complex marketing research. In fact, this company has an excellent lead generation and high demand largely due to the interest of the audience to the technology. Indeed, eye tracking allows you to check specific reactions to the aspects of your marketing, which differs from the traditional consumer surveys. It causes a great popularity of eye tracking technology in the marketing field and this trend is growing.

In conclusion, eye tracking technology is here to stay. The technology is being incorporated into our everyday lives so its acceptance seems to be widespread. As soon as developments and enhancements are taking place, eye tracking will continue to elevate the of using technology for automation of commonplace devices and detection of medical and/or psychological deviances (and that is just the beginning) The future of eye tracking technology is not limited to individual researcher's imagination, its practical application in our lives is boundless.