A simple algorithm for the offline recalibration of eye-tracking data through best-fitting linear transformation

Area of Science:

• Ophthalmology
• Cognitive Science
• Computer Science

Background:

• Eye-tracking experiments demand high accuracy and precision.
• Poor calibration and drift correction can compromise data integrity.
• Existing methods may not adequately address these challenges.

Purpose of the Study:

• To introduce a novel algorithm for offline correction of eye-tracking data.
• To enhance the accuracy and precision of eye-tracking measurements.
• To provide a practical implementation for researchers.

Main Methods:

• Developed an algorithm for offline correction of eye-tracking data.
• Employed a linear transformation of fixation coordinates.
• Minimized the distance between fixations and their closest stimulus.
• Implemented the algorithm in MATLAB.

Main Results:

• The algorithm demonstrated improved data quality under various conditions.
• Performance was evaluated using simulated and real eye-tracking data.
• Significant improvements were observed when a large number of fixations were used in the fitting process.

Conclusions:

• The proposed algorithm effectively corrects eye-tracking data, addressing calibration and drift issues.
• This method offers a valuable tool for researchers requiring precise eye-tracking data.
• The MATLAB implementation facilitates adoption and further research.