MAGE: A Multi-task Architecture for Gaze Estimation with an Efficient Calibration Module

TL;DR

MAGE is a multi-task neural network architecture that estimates 6-DoF gaze in 3D space, incorporating an efficient calibration module to adapt to individual differences, advancing gaze analysis in human-robot interaction.

Contribution

The paper introduces MAGE, a novel multi-task model with a calibration module for accurate 6-DoF gaze estimation applicable in real-world HRI scenarios.

Findings

Achieves state-of-the-art performance on multiple datasets.

Effectively adapts to individual eye variations.

Provides comprehensive 3D gaze information.

Abstract

Eye gaze can provide rich information on human psychological activities, and has garnered significant attention in the field of Human-Robot Interaction (HRI). However, existing gaze estimation methods merely predict either the gaze direction or the Point-of-Gaze (PoG) on the screen, failing to provide sufficient information for a comprehensive six Degree-of-Freedom (DoF) gaze analysis in 3D space. Moreover, the variations of eye shape and structure among individuals also impede the generalization capability of these methods. In this study, we propose MAGE, a Multi-task Architecture for Gaze Estimation with an efficient calibration module, to predict the 6-DoF gaze information that is applicable for the real-word HRI. Our basic model encodes both the directional and positional features from facial images, and predicts gaze results with dedicated information flow and multiple decoders. To reduce the impact of individual variations, we propose a novel calibration module, namely Easy-Calibration, to fine-tune the basic model with subject-specific data, which is efficient to implement without the need of a screen. Experimental results demonstrate that our method achieves state-of-the-art performance on the public MPIIFaceGaze, EYEDIAP, and our built IMRGaze datasets.