Saved in:
Bibliographic Details
Main Authors: Li, Zhenhao, Liu, Zheng, Lee, Seunghyun, Fadaeinejad, Amin, Yu, Yuanhao
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.26945
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908917821341696
author Li, Zhenhao
Liu, Zheng
Lee, Seunghyun
Fadaeinejad, Amin
Yu, Yuanhao
author_facet Li, Zhenhao
Liu, Zheng
Lee, Seunghyun
Fadaeinejad, Amin
Yu, Yuanhao
contents Appearance-based gaze estimation (AGE) has achieved remarkable performance in constrained settings, yet we reveal a significant generalization gap where existing AGE models often fail in practical, unconstrained scenarios, particularly those involving facial wearables and poor lighting conditions. We attribute this failure to two core factors: limited image diversity and inconsistent label fidelity across different datasets, especially along the pitch axis. To address these, we propose a robust AGE framework that enhances generalization without requiring additional human-annotated data. First, we expand the image manifold via an ensemble of augmentation techniques, including synthesis of eyeglasses, masks, and varied lighting. Second, to mitigate the impact of anisotropic inter-dataset label deviation, we reformulate gaze regression as a multi-task learning problem, incorporating multi-view supervised contrastive (SupCon) learning, discretized label classification, and eye-region segmentation as auxiliary objectives. To rigorously validate our approach, we curate new benchmark datasets designed to evaluate gaze robustness under challenging conditions, a dimension largely overlooked by existing evaluation protocols. Our MobileNet-based lightweight model achieves generalization performance competitive with the state-of-the-art (SOTA) UniGaze-H, while utilizing less than 1\% of its parameters, enabling high-fidelity, real-time gaze tracking on mobile devices.
format Preprint
id arxiv_https___arxiv_org_abs_2603_26945
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Real-time Appearance-based Gaze Estimation for Open Domains
Li, Zhenhao
Liu, Zheng
Lee, Seunghyun
Fadaeinejad, Amin
Yu, Yuanhao
Computer Vision and Pattern Recognition
Appearance-based gaze estimation (AGE) has achieved remarkable performance in constrained settings, yet we reveal a significant generalization gap where existing AGE models often fail in practical, unconstrained scenarios, particularly those involving facial wearables and poor lighting conditions. We attribute this failure to two core factors: limited image diversity and inconsistent label fidelity across different datasets, especially along the pitch axis. To address these, we propose a robust AGE framework that enhances generalization without requiring additional human-annotated data. First, we expand the image manifold via an ensemble of augmentation techniques, including synthesis of eyeglasses, masks, and varied lighting. Second, to mitigate the impact of anisotropic inter-dataset label deviation, we reformulate gaze regression as a multi-task learning problem, incorporating multi-view supervised contrastive (SupCon) learning, discretized label classification, and eye-region segmentation as auxiliary objectives. To rigorously validate our approach, we curate new benchmark datasets designed to evaluate gaze robustness under challenging conditions, a dimension largely overlooked by existing evaluation protocols. Our MobileNet-based lightweight model achieves generalization performance competitive with the state-of-the-art (SOTA) UniGaze-H, while utilizing less than 1\% of its parameters, enabling high-fidelity, real-time gaze tracking on mobile devices.
title Real-time Appearance-based Gaze Estimation for Open Domains
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.26945