Privacy-Friendly Synthetic Data for the Development of Face Morphing Attack Detectors
Abstract
The main question this work aims at answering is: "can morphing attack detection (MAD) solutions be successfully developed based on synthetic data?". Towards that, this work introduces the first synthetic-based MAD development dataset, namely the Synthetic Morphing Attack Detection Development dataset (SMDD). This dataset is utilized successfully to train three MAD backbones where it proved to lead to high MAD performance, even on completely unknown attack types. Additionally, an essential aspect of this work is the detailed legal analyses of the challenges of using and sharing real biometric data, rendering our proposed SMDD dataset extremely essential. The SMDD dataset, consisting of 30,000 attack and 50,000 bona fide samples, is publicly available for research purposes 1.
Cite
Text
Damer et al. "Privacy-Friendly Synthetic Data for the Development of Face Morphing Attack Detectors." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2022. doi:10.1109/CVPRW56347.2022.00167Markdown
[Damer et al. "Privacy-Friendly Synthetic Data for the Development of Face Morphing Attack Detectors." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2022.](https://mlanthology.org/cvprw/2022/damer2022cvprw-privacyfriendly/) doi:10.1109/CVPRW56347.2022.00167BibTeX
@inproceedings{damer2022cvprw-privacyfriendly,
title = {{Privacy-Friendly Synthetic Data for the Development of Face Morphing Attack Detectors}},
author = {Damer, Naser and López, César Augusto Fontanillo and Fang, Meiling and Spiller, Noémie and Pham, Minh Vu and Boutros, Fadi},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
year = {2022},
pages = {1605-1616},
doi = {10.1109/CVPRW56347.2022.00167},
url = {https://mlanthology.org/cvprw/2022/damer2022cvprw-privacyfriendly/}
}