Enhancing Appearance-Based Gaze Estimation Through Attention-Based Convolutional Neural Networks

Rawdha Karmi, Rawdha Karmi, Ines Rahmany, Ines Rahmany, Nawres Khlifa

2025

Abstract

Appearance-based gaze estimation is crucial for applications like assistive technology and human-computer interaction, but high accuracy is challenging due to complex gaze patterns and individual appearance variations. This paper proposes an Attention-Enhanced Convolutional Neural Network (AE-CNN) to address these challenges. By integrating attention submodules, AE-CNN improves feature extraction by focusing on the most relevant regions of input data. We evaluate AE-CNN using the ColumbiaGaze dataset and show that it surpasses previous methods, achieving a remarkable accuracy of 99.98%. This work advances gaze estimation by leveraging attention mechanisms to improve performance.

Paper Citation

in Harvard Style

Karmi R., Rahmany I. and Khlifa N. (2025). Enhancing Appearance-Based Gaze Estimation Through Attention-Based Convolutional Neural Networks. In Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART; ISBN 978-989-758-737-5, SciTePress, pages 15-23. DOI: 10.5220/0013055500003890

in Bibtex Style

@conference{icaart25,
author={Rawdha Karmi and Ines Rahmany and Nawres Khlifa},
title={Enhancing Appearance-Based Gaze Estimation Through Attention-Based Convolutional Neural Networks},
booktitle={Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART},
year={2025},
pages={15-23},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013055500003890},
isbn={978-989-758-737-5},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART
TI - Enhancing Appearance-Based Gaze Estimation Through Attention-Based Convolutional Neural Networks
SN - 978-989-758-737-5
AU - Karmi R.
AU - Rahmany I.
AU - Khlifa N.
PY - 2025
SP - 15
EP - 23
DO - 10.5220/0013055500003890
PB - SciTePress