Impact of Synthetic Images on Morphing Attack Detection Using a Siamese Network
This work addresses the challenge of enhancing security in biometric systems against morphing attacks, but it is incremental as it builds on existing Siamese network methods and suggests mixed training approaches.
The paper tackled the problem of improving Morphing Attack Detection (MAD) by evaluating the impact of synthetic images in training, finding that using EfficientNetB0 with real datasets achieved a lower error rate compared to state-of-the-art methods, while training only on synthetic images led to worse performance.
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.