Home / Current Issue / Paper 1712514
Eye Gaze Detection
Subject area: Science,Engineering and Technology · Area of research: Machine Learning
Abstract
Eye gaze detection is becoming an essential component in modern Human?Computer Interaction (HCI) due to its applications in assistive systems, driver monitoring, educational technologies, and immersive environments like VR/AR. Traditional gaze-tracking relies heavily on infrared sensors or specialized hardware, making such systems expensive and inaccessible for wide-scale deployment. This paper introduces GazeTrack, a real-time, lightweight, and cost-effective deep learning-based gaze detection system that operates solely using a standard webcam. By combining MediaPipe Face Mesh for extracting 468 facial landmarks with geometric analysis of the eye region, the system accurately classifies gaze directions such as Left, Right, Up, Down, and Center. The system maintains high accuracy (above 90%) and real-time performance (30?60 FPS) on standard hardware, making it highly suitable for low-cost real-world applications.
References
[1] Hansen, D. W., & Ji, Q. (2010). In the eye of the beholder: A survey of models for eyes and gaze. IEEE Transactions on Pattern Analysis and Machine Intelligence, 32(3), 478–500.
[2] Morimoto, C. H., & Mimica, M. R. (2005). Eye gaze tracking techniques for interactive applications. Computer Vision and Image Understanding, 98(1), 4–24.
[3] Yoo, D. H., Kim, J., Chung, I. J., & Chung, M. J. (2002). Non-intrusive eye gaze estimation using a single camera. Proceedings of the 2002 International Conference on Pattern Recognition.
[4] Bradski, G. (2000). The OpenCV Library. Dr. Dobb’s Journal: Software Tools for the Professional Programmer.
[5] Lugaresi, C., et al. (2019). MediaPipe: A Framework for Building Perception Pipelines. Google AI Research.
[6] Zhang, X., Sugano, Y., Fritz, M., & Bulling, A. (2015). Appearance-based gaze estimation in the wild. IEEE Conference on Computer Vision and Pattern Recognition (CVPR).
[7] Wood, E., et al. (2016). A 3D model of the human eye for gaze estimation. European Conference on Computer Vision (ECCV).
[8] Zhang, X., Sugano, Y., Fritz, M., & Bulling, A. (2017). MPIIGaze: Real-world dataset for robust gaze estimation. International Journal of Computer Vision.
[9] Krafka, K., et al. (2016). Eye Tracking for Everyone. IEEE CVPR (GazeCapture Dataset).
[10] Duchowski, A. T. (2007). Eye Tracking Methodology: Theory and Practice. Springer.
How to cite this paper
@article{1712514,
author = {Dhanushree C P, Nisarga N P, Manoj Kumar C S, Niba Mehak, Abdul Rahman},
title = {Eye Gaze Detection},
journal = {Iconic Research And Engineering Journals},
year = {2025},
volume = {9},
number = {5},
pages = {2292-2295},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1712514.pdf},
abstract = {Eye gaze detection is becoming an essential component in modern Human?Computer Interaction (HCI) due to its applications in assistive systems, driver monitoring, educational technologies, and immersive environments like VR/AR. Traditional gaze-tracking relies heavily on infrared sensors or specialized hardware, making such systems expensive and inaccessible for wide-scale deployment. This paper introduces GazeTrack, a real-time, lightweight, and cost-effective deep learning-based gaze detection system that operates solely using a standard webcam. By combining MediaPipe Face Mesh for extracting 468 facial landmarks with geometric analysis of the eye region, the system accurately classifies gaze directions such as Left, Right, Up, Down, and Center. The system maintains high accuracy (above 90%) and real-time performance (30?60 FPS) on standard hardware, making it highly suitable for low-cost real-world applications.},
month = {November},
doi = {https://doi.org/10.64388/IREV9I5-1712514}
}