Abstract
Coupled with deep learning technology, a field of gaze estimation has shown remarkable advances in recent decades. These advances however relied on the large amount of computations and resources due to the complex deep learning architecture or the huge volume of data, respectively. In this paper, we propose a resource efficient approach with ensemble loss function to improve the gaze estimation performance. Since eye gaze estimation is initially a regression problem in a broad sense, we employed the ensemble technique of regression loss functions in the pursuit of estimating gaze coordinates on a higher precision, instead of using additional deep layers or much data. Preliminary experiments on MPIIGaze data showed the improved performance compared to state-of-the-art models, the mean average error of 3.7887 cm which is 10.2%, 50.7%, 42.9% better than that of AFF-Net, Itracker, GazeNet, respectively.
S.H. Kim and S.G. Lee—Are co-first authors and contributed equally.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Park, S., et al.: Few-shot adaptive gaze estimation. In: 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Korea (South) (2019)
Krafka, K., et al.: Eye tracking for everyone. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2176–2184 (2016)
He, J., et al.: On-device few-shot personalization for real-time gaze estimation. In: 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), pp. 1149–1158 (2019)
Hajiabadi, H., Monsefi, R., Yazdi, H.S.: RELF: robust regression extended with ensemble loss function. Appl. Intell. 49(4), 1437–1450 (2018). https://doi.org/10.1007/s10489-018-1341-9
Bao, Y., Cheng, Y., Liu, Y., Lu, F.: Adaptive feature fusion network for gaze tracking in mobile tablets. In: 2020 25th International Conference on Pattern Recognition (ICPR), pp. 9936–9943 (2021)
Abdelrahman, A., Hempel, T., Khalifa, A., Al-Hamadi, A.: L2CS-Net: fine-grained gaze estimation in unconstrained environments (2022)
Hu, J., Shen, L., Sun, G.: Squeeze-and-excitation networks. In: 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 7132–7141 (2018)
Zhang, X., Sugano, Y., Fritz, M., Bulling, A.: MPIIGaze: real-world dataset and deep appearance-based gaze estimation. IEEE Trans. Pattern Anal. Mach. Intell. 41(01), 162–175 (2019)
Cheng, Y., Wang, H., Bao, Y., Lu, F: Appearance-based gaze estimation with deep learning: a review and benchmark. ArXiv (2021)
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Kim, S.H., Lee, S.G., Lee, J.H., Lee, E.C. (2023). Improving Gaze Estimation Performance Using Ensemble Loss Function. In: Zaynidinov, H., Singh, M., Tiwary, U.S., Singh, D. (eds) Intelligent Human Computer Interaction. IHCI 2022. Lecture Notes in Computer Science, vol 13741. Springer, Cham. https://doi.org/10.1007/978-3-031-27199-1_51
Download citation
DOI: https://doi.org/10.1007/978-3-031-27199-1_51
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-27198-4
Online ISBN: 978-3-031-27199-1
eBook Packages: Computer ScienceComputer Science (R0)