GazeBaseVR: 5,020 Binocular Eye-Tracking Recordings from 407 Participants

A 26-month longitudinal study collected 5,020 binocular eye-tracking recordings at 250 Hz from 407 college-aged participants using a VR headset. Participants performed five tasks including vergence, smooth pursuit, video viewing, reading, and saccade tasks. The dataset, created by Dillon Lohr, is suitable for research on eye movement biometrics and fairness in VR.

Use Cases

• Developing biometric identification models based on longitudinal eye movement patterns.
• Studying the effects of COVID-19 infection and recovery on oculomotor control using data from 11 participants.
• Training and evaluating smooth pursuit or saccade detection algorithms for VR applications.
• Investigating fairness in biometric systems using the provided participant details.

Strengths

• Large-scale longitudinal design with 5,020 recordings from 407 participants over 26 months.
• High-frequency 250 Hz binocular data collection suitable for detailed oculomotor analysis.
• Includes data from participants recorded with other devices, enabling cross-device validation studies.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file formats are unknown, which may limit suitability assessment.
• Data reflects a specific demographic bias, being collected from a college-aged population.

Provenance

Source

Dillon Lohr

Collection Method

Collected using an eye-tracking enabled virtual reality headset during five structured tasks.

Time Range

Longitudinal data collected over a 26-month period.