The Effect of Real-time Headbox Adjustments on Data Quality

Area of Science:

• Computer Vision
• Human-Computer Interaction
• Biomedical Engineering

Background:

• High framerates are crucial for accurate eye-tracking.
• Traditional methods often face limitations in data processing speed.
• Optimizing camera sensor usage is key to improving eye-tracking system performance.

Purpose of the Study:

• To investigate the impact of real-time headbox adjustments on eye-tracking parameters.
• To evaluate the trade-offs between framerate and accuracy with dynamic recording window adjustments.
• To determine optimal headbox adjustment frequency for enhanced eye-tracking data acquisition.

Main Methods:

• Implemented a dynamic recording window technique based on Tobii patent.
• Adjusted the size and position of the recording window in real-time to track eye movements.
• Measured eye-tracking performance metrics including framerate, accuracy, precision, and latency.
• Analyzed the effect of headbox adjustment frequency (1-2 adjustments/sec) on these parameters.

Main Results:

• Higher framerates were achieved by reducing the recording window size.
• One to two headbox adjustments per second were found to be tolerable without significant degradation of key eye-tracking parameters.
• Increasing the recording window size reduced the impact of adjustments but lowered the achievable framerate.

Conclusions:

• Dynamic adjustment of the recording window is an effective method for increasing eye-tracking camera framerates.
• A balance between headbox adjustment frequency and recording window size can optimize eye-tracking performance for various applications.
• The findings support the feasibility of achieving high-speed, accurate eye-tracking while allowing natural participant movement.