RealEye Accuracy on Computers
We conducted a study to evaluate the accuracy of the RealEye webcam eye-tracking on computers. The data were collected from various participants in diverse environments, ensuring the results are both realistic and robust. Based on our findings, the following conclusions can be drawn:
Overall Accuracy: The average Accuracy ranged between 115 and 125 pixels, depending on the quality of the data. For data categorized as at least "Good", the average Accuracy = 123 pixels. It is anticipated that Accuracy could be improved if measurements were taken immediately after Calibration.
Best overall Accuracy: 61 pixels.
Accuracy by screen area: Variations in Accuracy were observed across different parts of the screen, with a range from 99 to 154 pixels. The central-bottom part of the screen yielded the best Accuracy, while the bottom-right corner exhibited the lowest accuracy.
Overall Precision: The average Precision ranged between 74 and 77 pixels, depending on the quality of the data. For data categorized as at least "Good", the average Precision = 76 pixels.
Best overall Precision: 29 pixels.
Precision by screen area: Variations in Precision were observed across different parts of the screen, with a range from 62 to 90 pixels. The Precision was best in the left-central-to-bottom part of the screen, and worst in the top-right part of the screen.
The worst results for both Accuracy and Precision were noted in the corners of the screen.
Participant Performance: 27% of participants achieved an accuracy of up to 100 pixels, while 73% achieved an accuracy of up to 150 pixels.
Read more here: RealEye Technical Whitepaper.
Please remember that the better hardware/software your participant has, the more accurate results you'll get. Check the recommended requirements.
NOTE:
In the following paper:
Katarzyna Wisiecka, Krzysztof Krejtz, Izabela Krejtz, Damian Sromek, Adam Cellary, Beata Lewandowska, and Andrew Duchowski. 2022. Comparison of Webcam and Remote Eye Tracking. In 2022 Symposium on Eye Tracking Research and Applications (ETRA' 22). Association for Computing Machinery, New York, NY, USA, Article 32, 1–7. https://doi.org/10.1145/3517031.3529615
we compared measurement error and validity of the RealEye webcam solution to remote (the GP3) eye-tracking and an integrated method in point detection tasks. We measured the error defined as a distance (in pixels) between the center of the AOI and the position of the participant's fixation. In laboratory conditions, this distance was:
webcam - 45.1,
remote - 34.2,
integrated condition - 27.70.