cvrl / deformirisnet

Code for DeformIrisNet: An Identity-Preserving Model of Iris Texture Deformation

To run the network to dilate a single pupil image, use single_image.py as follows:

python single_image.py <weight_path> <small_pupil_image_path> <big_pupil_mask_image_path>

The weight is the .pth file provided here: https://notredame.box.com/v/deformirisnetweight . A large pupil image will be saved as dilated.png, play around with the mask and see the output.

The paper can be found here:

https://openaccess.thecvf.com/content/WACV2023/papers/Khan_DeformIrisNet_An_Identity-Preserving_Model_of_Iris_Texture_Deformation_WACV_2023_paper.pdf

https://arxiv.org/abs/2207.08980

Bibtex:

@inproceedings{khan2023deformirisnet,
    title = {DeformIrisNet: An Identity-Preserving Model of Iris Texture Deformation},
    author = {Khan, Siamul Karim and Tinsley, Patrick and Czajka, Adam},
    booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision},
    pages = {900--908},
    year = {2023}
}