The minimum AOI size is set to ensure that the data output based off the accuracy of Sticky´s Autogazer webcam eye tracking algorithm, is valid. The area has to be large enough for us to capture the data to provide you with the most accurate results. Accuracy - normally defined as the average difference between the real gaze position and the measured gaze position - is one of the most important metrics to the performance of an eye tracking solution. A system with good accuracy will provide more valid data as it is able to truthfully describe the location of a respondent’s gaze on a screen. It is therefore important to measure a system’s accuracy in order to evaluate the relevance of the eye tracking data the solution produces.
Through our validation experiments on internal true data stimuli, we have determined that Autogazer’s average gaze error in a real-world (Non-lab) environment is 1.6-1.8 degrees (~5% of the screen width and 7% of the screen height) on a laptop. In live usage we assume that we need to be able to register at least 90% of all respondents gazing at a particular area of interest and we need to capture at least half of the theoretical maximal time spent on that area. To achieve this with Autogazer, the area of interest should be 200x200 pixels or more. This is more than accurate enough to differentiate what people are looking at for elements of a web page, package, image, advertising or other stimulus.
As part of our true data stimuli testing we used different face images to test the accuracy. Below you can see the difference in using the 100x100, 200x200 or 300x300 pixel areas of interest.