Pose Tracking for Automated Video-Oculography Using Handheld, Mobile Devices 2413 - 0872

Video-oculography (VOG) enables the objective measurement of eye movements for clinically-relevant tasks, but wide-spread application via consumer devices requires sub-mm and sub-degree head pose tracking to compensate for uncontrolled head and camera motion. We consider approaches: Apple's ARKit, which fuses TrueDepth IR and RGB but is restricted to iPhones with Face ID hardware, and Pixel3DMM, a monocular RGB-only method that could generalize to any camera but has not been validated against depth-sensor ground truth. Head pose from both trackers was aligned frame-by-frame and split into head-still and head-moving portions. We measured peak-to-peak change at each head-movement event (E→F vertical, G→H horizontal), drift as the endpoint difference across the head-still segment, and saccade response as mean peak-to-peak across detected vertical and horizontal saccade events.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-16
DOI
https://doi.org/10.5281/zenodo.24294756
Primary Topic
Gaze Tracking and Assistive Technology
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Pose Tracking for Automated Video-Oculography Using Handheld, Mobile Devices 2413 - 0872

Yue Wu, Pranav Narsipur, Yelena Bagdasarova
1 links
Zenodo (CERN European Organization for Nuclear Research)
Gaze Tracking and Assistive Technology
article

Pose Tracking for Automated Video-Oculography Using Handheld, Mobile Devices 2413 - 0872

Yue Wu, Pranav Narsipur, Yelena Bagdasarova
article en

Abstract

Video-oculography (VOG) enables the objective measurement of eye movements for clinically-relevant tasks, but wide-spread application via consumer devices requires sub-mm and sub-degree head pose tracking to compensate for uncontrolled head and camera motion. We consider approaches: Apple's ARKit, which fuses TrueDepth IR and RGB but is restricted to iPhones with Face ID hardware, and Pixel3DMM, a monocular RGB-only method that could generalize to any camera but has not been validated against depth-sensor ground truth. Head pose from both trackers was aligned frame-by-frame and split into head-still and head-moving portions. We measured peak-to-peak change at each head-movement event (E→F vertical, G→H horizontal), drift as the endpoint difference across the head-still segment, and saccade response as mean peak-to-peak across detected vertical and horizontal saccade events.

Zenodo (CERN European Organization for Nuclear Research)
University of Washington (US)
Gaze Tracking and Assistive Technology
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon