Is Geometry Enough? An Evaluation of Landmark-Based Gaze Estimation

Appearance-based gaze estimation frequently relies on deep Convolutional Neural Networks (CNNs). These models are accurate, but computationally expensive and act as ‘‘black boxes’’, offering little interpretability. Geometric methods based on facial landmarks are lightweight alternatives. However, their performance limits and generalization capabilities remain underexplored in modern benchmarks. In this study, we conduct a comprehensive evaluation of the landmark-based gaze estimation. We introduce a standardized pipeline to extract and normalize landmarks from three large-scale datasets (Gaze360, ETH-XGaze, and GazeGene) and train lightweight regression models, specifically extreme gradient boosted trees and two neural architectures: a holistic multilayer perceptron (MLP) and a Siamese MLP designed to capture binocular geometry. We find that landmark-based models exhibit lower performance in the within-domain evaluation, likely due to noise introduced into the datasets by the landmark detector. Never-theless, in the cross-domain evaluation, the proposed MLP architectures show generalization capabilities comparable to those of the ResNet18 baselines. These findings suggest that sparse geometric features encode sufficient information for robust gaze estimation, paving the way for efficient, interpretable, and privacy-friendly edge applications. The source code and generated landmark-based datasets are available at: https://github.com/daniele-agostinelli/LandmarkGaze.git.