Sightcorp logo

Eye Tracking Software

Everything about Eye Tracking Software

     

Eye-Tracking Software: What is it and how does it work?

With eye-tracking software, you can find out where users focus their visual attention when browsing a website, using a product, browsing a store, or viewing a digital ad. You can then use this data to tailor your content, adjust layouts, improve the user experience, and ultimately increase ROI.

What is eye-tracking technology?

Eye tracking technology is a research tool that monitors and records eye behavior such as pupil movement and dilation in response to a stimulus. The stimulus could be a website, a digital ad, a product design, or something else that you would like to conduct market research or user testing on.

How does eye-tracking software work?

Eye-tracking software can work in various ways. Some software works with infrared eye-tracking cameras, while some software can also work with regular webcams (which allow for online eye tracking). The cameras capture images of the eyes while they are focused on the stimuli, and send the data through to the eye-tracking software, which processes the images. This image processing is usually based on AI. Once the AI algorithms have processed the images, you will usually be able to see the data output in the form of heatmaps and other representations of the users’ attention.

 

Request Market Research

The questions you can answer using this data include:

  • Which areas received the most attention?
  • How much time did people spend looking at a particular area?
  • In which order did the different areas receive attention? Which direction did the user’s attention flow in?

Once you have these answers, you can make the relevant adjustments to improve the user experience.

Eye tracking and face analysis

At Sightcorp, we have in the past used eye-tracking technology in combination with emotion recognition and other elements of face analysis to offer comprehensive market research solutions. By tracking eye movements and attention time in conjunction with facial analysis, it is possible to gain a clear picture of how a user or customer responds to a website, product, store layout, or digital ad, and to identify areas for improvement.

While we are no longer offering eye-tracking solutions, we now offer a deep learning-based facial analysis tool, DeepSight SDK, which can tell you the following about your audience:

  • Age
  • Gender
  • Audience size
  • Attention time (e.g. for how long, on average, did each person spend looking at an ad?)
  • Headpose

 

In DeepSight, headpose estimation replaces eye tracking, which was previously quite difficult to implement in real-life situations since it requires calibration. While headpose detection provides less detail than eye tracking, it still enables you to determine gaze direction, while being easier to use in real-life environments. Eye tracking is better suited to market research studies conducted with individuals, where the user is situated in a stationary position near the camera. Headpose estimation, on the other hand, is great for crowd analysis, where people are moving around, at varying distances from the camera.

Click here to learn more about DeepSight SDK

Also read:

   

Technical Specifications

The table below shows how FaceMatch SDK performs on the Labelled Faces in the Wild (LFW) dataset:

FPRTPRThreshold (Inverse of distance)
0.10.99900 ±0.002130.55448
0.010.99667 ±0.005370.59791
0.0010.99367 ±0.006050.62989

FPR = False Positive Rate
TPR = True Positive Rate

These results are an indication only and are based on the specific dataset Labelled Faces in the Wild. Customers can expect similar performance, with possible variations due to hardware and the availability of annotated data.