A Calibration-Free Multi-Expert System for Automated Diopter Measurement: Overcoming Noise-Induced Binocular Inconsistency
IEEE Access · 2026 · 14 · 71762-71775
DOI: 10.1109/access.2026.3691790 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Thimaporn Phetkaew und 2 weiteren Autor:innen. Die Arbeit nennt 36 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Walailak University, Reinventing University Program through the Ministry of Higher Education, Science, Research and Innovation.
Abstract (Originalsprache)
Estimating ocular alignment in unconstrained settings remains challenging because Noise-Induced Binocular Inconsistency (NIBI), caused by blinking, head motion, and tracking noise, can produce temporary inter-ocular inconsistencies. Existing frame-based or classification-based methods often cannot provide reliable continuous measurements for clinical use. In this study, we introduce a calibration-free deep multi-expert architecture for real-time prism diopter estimation from short monocular video sequences. Unlike approaches that treat ocular alignment as a static image-, feature-, or frame-wise task, the proposed framework formulates it as a temporal regression problem to better handle dynamic image variations. The framework integrates three complementary components: a Geometric Expert, termed the Living Ruler, which extracts iris-normalized and scale-invariant features; a Trajectory Expert, which encodes temporal eye movement dynamics together with deep image features; and a Transformer-based Stability Supervisor, which estimates frame-wise reliability scores for reliability-aware temporal fusion by emphasizing stable observations and suppressing unreliable ones. Because publicly available datasets with prism diopter annotations are limited, geometry-based reference deviations were simulated using the real human eye geometry derived from the Columbia Gaze Dataset. Experimental results show that the proposed method achieves high accuracy, with a mean absolute error of $1.3038\Delta $ and an $R^{2}$ of 0.9963 over the range of 0– $50\Delta $ , while maintaining strong temporal stability. Evaluation on real-world data without fine-tuning further demonstrated generalization across appearance, head pose, and imaging variations. These findings suggest that continuous calibration-free ocular deviation estimation from standard video input is feasible for tele-screening and longitudinal monitoring.
Förderung
- Walailak University
- Reinventing University Program through the Ministry of Higher Education, Science, Research and Innovation
Zitieren
APA
Phetkaew, T., Kurdthongmee, W., & Duh, W. (2026). A Calibration-Free Multi-Expert System for Automated Diopter Measurement: Overcoming Noise-Induced Binocular Inconsistency. IEEE Access, 14, 71762-71775. https://doi.org/10.1109/access.2026.3691790
Vancouver
Phetkaew T, Kurdthongmee W, Duh W. A Calibration-Free Multi-Expert System for Automated Diopter Measurement: Overcoming Noise-Induced Binocular Inconsistency. IEEE Access. 2026;14:71762-71775. doi:10.1109/access.2026.3691790
BibTeX
@article{Phetkaew2026,
author = {Phetkaew, Thimaporn and Kurdthongmee, Wattanapong and Duh, Wei-Jou},
title = {A Calibration-Free Multi-Expert System for Automated Diopter Measurement: Overcoming Noise-Induced Binocular Inconsistency},
journal = {IEEE Access},
year = {2026},
volume = {14},
pages = {71762--71775},
doi = {10.1109/access.2026.3691790}
}
EUVIMED