High-frequency near-eye ground truth for event-based eye tracking

TL;DR

This paper introduces an improved event-based eye-tracking dataset with semi-automatic annotations at 200Hz, addressing the scarcity of eye-level annotated data for algorithm validation and deep learning training.

Contribution

It presents a semi-automatic annotation pipeline and provides a new high-frequency dataset for event-based eye tracking research.

Findings

Enhanced dataset with pupil detection annotations at 200Hz

Facilitates algorithm validation and deep learning training

Addresses data scarcity in event-based eye tracking

Abstract

Event-based eye tracking is a promising solution for efficient and low-power eye tracking in smart eyewear technologies. However, the novelty of event-based sensors has resulted in a limited number of available datasets, particularly those with eye-level annotations, crucial for algorithm validation and deep-learning training. This paper addresses this gap by presenting an improved version of a popular event-based eye-tracking dataset. We introduce a semi-automatic annotation pipeline specifically designed for event-based data annotation. Additionally, we provide the scientific community with the computed annotations for pupil detection at 200Hz.