Eye Semantic Segmentation with a Lightweight Model

AR/VR Eye Semantic Segmentation with Open Eye Dataset [1] in OpenEDS Semantic Segmentation Challenge 2019.

• Best mIoU: 0.9491
• Number of trainable parameters: 104728

V. T. Huynh, S. Kim, G. Lee and H. Yang, "Eye Semantic Segmentation with A Lightweight Model," 2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW), Seoul, Korea (South), 2019, pp. 3694-3697, doi: 10.1109/ICCVW.2019.00457.

@INPROCEEDINGS{9022251,  
    author={V. T. {Huynh} and S. {Kim} and G. {Lee} and H. {Yang}},  
    booktitle={2019 IEEE/CVF International Conference on Computer Vision Workshop (ICCVW)},   
    title={Eye Semantic Segmentation with A Lightweight Model},   
    year={2019},  
    pages={3694-3697},
}

V. T. Huynh, H. Yang, G. Lee and S. Kim, "Semantic Segmentation of the Eye With a Lightweight Deep Network and Shape Correction," in IEEE Access, vol. 8, pp. 131967-131974, 2020, doi: 10.1109/ACCESS.2020.3010011.

@ARTICLE{9143078,  
    author={V. T. {Huynh} and H. {Yang} and G. {Lee} and S. {Kim}},  
    journal={IEEE Access},   
    title={Semantic Segmentation of the Eye With a Lightweight Deep Network and Shape Correction},   
    year={2020},  
    volume={8},  
    pages={131967-131974},
}

[1] Garbin, Stephan J., Yiru Shen, Immo Schuetz, Robert Cavin, Gregory Hughes, and Sachin S. Talathi. "OpenEDS: Open Eye Dataset." arXiv preprint arXiv:1905.03702 (2019).