Pose-Robust Calibration Strategy for Point-of-Gaze Estimation on Mobile Phones

TL;DR

This paper presents a pose-robust calibration method for point-of-gaze estimation on mobile phones, using a new benchmark and a dynamic calibration strategy that improves accuracy across head poses.

Contribution

It introduces a new benchmark, MobilePoG, and proposes a dynamic calibration strategy that enhances gaze estimation robustness to head pose variations.

Findings

Wider head pose variation during calibration improves accuracy.

Dynamic calibration reduces sensitivity to head pose changes.

The proposed method outperforms conventional calibration strategies.

Abstract

Although appearance-based point-of-gaze (PoG) estimation has improved, the estimators still struggle to generalize across individuals due to personal differences. Therefore, person-specific calibration is required for accurate PoG estimation. However, calibrated PoG estimators are often sensitive to head pose variations. To address this, we investigate the key factors influencing calibrated estimators and explore pose-robust calibration strategies. Specifically, we first construct a benchmark, MobilePoG, which includes facial images from 32 individuals focusing on designated points under either fixed or continuously changing head poses. Using this benchmark, we systematically analyze how the diversity of calibration points and head poses influences estimation accuracy. Our experiments show that introducing a wider range of head poses during calibration improves the estimator's ability to handle pose variation. Building on this insight, we propose a dynamic calibration strategy in which users fixate on calibration points while moving their phones. This strategy naturally introduces head pose variation during a user-friendly and efficient calibration process, ultimately producing a better calibrated PoG estimator that is less sensitive to head pose variations than those using conventional calibration strategies. Codes and datasets are available at our project page.