EEGMobile: Enhancing Speed and Accuracy in EEG-Based Gaze Prediction with Advanced Mobile Architectures

TL;DR

This paper introduces a lightweight EEG-based gaze prediction model using MobileViT and Knowledge Distillation, achieving near state-of-the-art accuracy with significantly improved speed and reduced size for mobile applications.

Contribution

The study presents a novel mobile-friendly EEG regression model that is faster, smaller, and nearly as accurate as the previous SOTA, enabling practical real-world BCI applications.

Findings

Model is 33% faster than SOTA

Model is 60% smaller in size

Achieves 97% of SOTA accuracy

Abstract

Electroencephalography (EEG) analysis is an important domain in the realm of Brain-Computer Interface (BCI) research. To ensure BCI devices are capable of providing practical applications in the real world, brain signal processing techniques must be fast, accurate, and resource-conscious to deliver low-latency neural analytics. This study presents a model that leverages a pre-trained MobileViT alongside Knowledge Distillation (KD) for EEG regression tasks. Our results showcase that this model is capable of performing at a level comparable (only 3% lower) to the previous State-Of-The-Art (SOTA) on the EEGEyeNet Absolute Position Task while being 33% faster and 60% smaller. Our research presents a cost-effective model applicable to resource-constrained devices and contributes to expanding future research on lightweight, mobile-friendly models for EEG regression.