Offset Calibration for Appearance-Based Gaze Estimation via Gaze Decomposition
This work addresses accuracy issues in gaze tracking for applications like human-computer interaction, offering both calibration-free and calibrated solutions, but it is incremental as it builds on existing gaze estimation methods.
The paper tackled the problem of limited accuracy in subject-independent appearance-based gaze estimation due to individual variations by proposing a gaze decomposition method and a single gaze point calibration method, resulting in up to 10.0% improvement in calibration-free estimation and up to 35.6% error reduction with calibration.
Appearance-based gaze estimation provides relatively unconstrained gaze tracking. However, subject-independent models achieve limited accuracy partly due to individual variations. To improve estimation, we propose a novel gaze decomposition method and a single gaze point calibration method, motivated by our finding that the inter-subject squared bias exceeds the intra-subject variance for a subject-independent estimator. We decompose the gaze angle into a subject-dependent bias term and a subject-independent term between the gaze angle and the bias. The subject-independent term is estimated by a deep convolutional network. For calibration-free tracking, we set the subject-dependent bias term to zero. For single gaze point calibration, we estimate the bias from a few images taken as the subject gazes at a point. Experiments on three datasets indicate that as a calibration-free estimator, the proposed method outperforms the state-of-the-art methods by up to $10.0\%$. The proposed calibration method is robust and reduces estimation error significantly (up to $35.6\%$), achieving state-of-the-art performance for appearance-based eye trackers with calibration.