FLAME: Facial Landmark Heatmap Activated Multimodal Gaze Estimation

TL;DR

This paper introduces FLAME, a new multimodal approach that uses facial landmark heatmaps to improve 3D gaze estimation accuracy without requiring person-specific calibration, showing significant performance gains.

Contribution

The paper presents FLAME, a novel method combining eye landmark heatmaps with multimodal data for calibration-free, precise 3D gaze estimation.

Findings

Approximately 10% performance improvement on ColumbiaGaze and EYEDIAP datasets.

Validation of the effectiveness through ablation studies.

Demonstrates competitive accuracy without person-specific calibration.

Abstract

3D gaze estimation is about predicting the line of sight of a person in 3D space. Person-independent models for the same lack precision due to anatomical differences of subjects, whereas person-specific calibrated techniques add strict constraints on scalability. To overcome these issues, we propose a novel technique, Facial Landmark Heatmap Activated Multimodal Gaze Estimation (FLAME), as a way of combining eye anatomical information using eye landmark heatmaps to obtain precise gaze estimation without any person-specific calibration. Our evaluation demonstrates a competitive performance of about 10% improvement on benchmark datasets ColumbiaGaze and EYEDIAP. We also conduct an ablation study to validate our method.