Effect of Kernel Size on CNN-Vision-Transformer-Based Gaze Prediction Using Electroencephalography Data

TL;DR

This paper introduces a novel EEG-based gaze prediction algorithm that improves accuracy and reduces training time compared to existing methods, offering a promising alternative to video-based eye-tracking.

Contribution

The study presents a new EEG-based gaze prediction method with optimized kernel sizes, achieving better accuracy and efficiency than prior approaches.

Findings

Root mean-squared-error reduced to 53.06 mm

Training time decreased to less than 33% of original

Source code publicly available

Abstract

In this paper, we present an algorithm of gaze prediction from Electroencephalography (EEG) data. EEG-based gaze prediction is a new research topic that can serve as an alternative to traditional video-based eye-tracking. Compared to the existing state-of-the-art (SOTA) method, we improved the root mean-squared-error of EEG-based gaze prediction to 53.06 millimeters, while reducing the training time to less than 33% of its original duration. Our source code can be found at https://github.com/AmCh-Q/CSCI6907Project