This paper evaluated the impact of synthetic images on Morphing Attack Detection (MAD) using a Siamese network with a semi-hard-loss function. Intra and cross-dataset evaluations were performed to measure synthetic image generalisation capabilities using a cross-dataset for evaluation. Three different pre-trained networks were used as feature extractors from traditional MobileNetV2, MobileNetV3 and EfficientNetB0. Our results show that MAD trained on EfficientNetB0 from FERET, FRGCv2, and FRLL can reach a lower error rate in comparison with SOTA. Conversely, worse performances were reached when the system was trained only with synthetic images. A mixed approach (synthetic + digital) database may help to improve MAD and reduce the error rate. This fact shows that we still need to keep going with our efforts to include synthetic images in the training process.

翻译：本文评估了使用半难损失函数训练的孪生网络进行形态攻击检测时，合成图像对其性能的影响。通过内部数据集与跨数据集评估，衡量了合成图像的泛化能力，其中跨数据集用于评估。实验采用三种不同的预训练网络作为特征提取器：传统的MobileNetV2、MobileNetV3和EfficientNetB0。结果表明，基于FERET、FRGCv2和FRLL数据集训练的EfficientNetB0形态攻击检测模型，其错误率低于现有最优方法。相反，仅使用合成图像训练的系统表现较差。采用混合（合成+数字）数据库的方法有助于提升检测性能并降低错误率。这一事实表明，我们仍需持续努力将合成图像纳入训练过程。