Evaluating Cross-Subject and Cross-Device Consistency in Visual Fixation Prediction
🤖 AI Summary
Embedded eye-trackers in smart glasses face challenges in cross-device and cross-subject consistency for visual attention modeling and neural prosthetics. Method: We conducted a nine-participant subjective experiment using the MIT1003 dataset to systematically quantify the cross-device transferability of mean fixation distributions from embedded eyewear to high-end standalone eye-trackers—first such evaluation. Consistency was assessed at both individual and group levels using AUC, NSS, and CC metrics. Results: Mean fixation distributions exhibit robust cross-device transferability under simple image stimuli; however, individual fixation patterns show significantly limited generalizability across devices, with low individual-to-group consistency. This reveals a critical bottleneck in deploying embedded eye-tracking for personalized neural interfaces. Moreover, the strong group-level consistency provides empirical support for leveraging population priors—rather than subject-specific calibration—in neural prosthetic applications.
Application Category
📝 Abstract
Understanding cross-subject and cross-device consistency in visual fixation prediction is essential for advancing eye-tracking applications, including visual attention modeling and neuroprosthetics. This study evaluates fixation consistency using an embedded eye tracker integrated into regular-sized glasses, comparing its performance with high-end standalone eye-tracking systems. Nine participants viewed 300 images from the MIT1003 dataset in subjective experiments, allowing us to analyze cross-device and cross-subject variations in fixation patterns with various evaluation metrics. Our findings indicate that average visual fixations can be reliably transferred across devices for relatively simple stimuli. However, individual-to-average consistency remains weak, highlighting the challenges of predicting individual fixations across devices. These results provide an empirical foundation for leveraging predicted average visual fixation data to enhance neuroprosthetic applications.
Problem
Research questions and friction points this paper is trying to address.
Evaluates cross-subject fixation consistency
Compares embedded vs. standalone eye trackers
Assesses device transferability of fixation predictions
Innovation
Methods, ideas, or system contributions that make the work stand out.
Embedded eye tracker glasses
Cross-device fixation consistency
MIT1003 dataset analysis
🔎 Similar Papers
No similar papers found.
