Labeled pupils in the wild: A dataset for studying pupil detection in unconstrained environments

TL;DR

This paper introduces LPW, a comprehensive dataset of high-quality eye videos capturing diverse real-world conditions to advance and evaluate pupil detection algorithms.

Contribution

The paper presents LPW, a new dataset with diverse conditions for developing and benchmarking pupil detection algorithms in unconstrained environments.

Findings

Benchmarking of five state-of-the-art algorithms on LPW

Analysis of factors affecting pupil detection accuracy

Insights into challenges for robust pupil detection

Abstract

We present labelled pupils in the wild (LPW), a novel dataset of 66 high-quality, high-speed eye region videos for the development and evaluation of pupil detection algorithms. The videos in our dataset were recorded from 22 participants in everyday locations at about 95 FPS using a state-of-the-art dark-pupil head-mounted eye tracker. They cover people with different ethnicities, a diverse set of everyday indoor and outdoor illumination environments, as well as natural gaze direction distributions. The dataset also includes participants wearing glasses, contact lenses, as well as make-up. We benchmark five state-of-the-art pupil detection algorithms on our dataset with respect to robustness and accuracy. We further study the influence of image resolution, vision aids, as well as recording location (indoor, outdoor) on pupil detection performance. Our evaluations provide valuable insights into the general pupil detection problem and allow us to identify key challenges for robust pupil detection on head-mounted eye trackers.