Gustavo Botelho de Souza

Johnatan S. Oliveira, Gustavo B. Souza, Anderson R. Rocha et al.

Ensuring the security of transactions is currently one of the major challenges that banking systems deal with. The usage of face for biometric authentication of users is attracting large investments from banks worldwide due to its convenience and acceptability by people, especially in cross-domain scenarios, in which facial images from ID documents are compared with digital self-portraits (selfies) for the automated opening of new checking accounts, e.g, or financial transactions authorization. Actually, the comparison of selfies and IDs has also been applied in another wide variety of tasks nowadays, such as automated immigration control. The major difficulty in such process consists in attenuating the differences between the facial images compared given their different domains. In this work, in addition to collecting a large cross-domain face dataset, with 27,002 real facial images of selfies and ID documents (13,501 subjects) captured from the databases of the major public Brazilian bank, we propose a novel architecture for such cross-domain matching problem based on deep features extracted by two well-referenced Convolutional Neural Networks (CNN). Results obtained on the dataset collected, called FaceBank, with accuracy rates higher than 93%, demonstrate the robustness of the proposed approach to the cross-domain face matching problem and its feasible application in real banking security systems.

Gustavo Botelho de Souza, João Paulo Papa, Aparecido Nilceu Marana

Biometrics emerged as a robust solution for security systems. However, given the dissemination of biometric applications, criminals are developing techniques to circumvent them by simulating physical or behavioral traits of legal users (spoofing attacks). Despite face being a promising characteristic due to its universality, acceptability and presence of cameras almost everywhere, face recognition systems are extremely vulnerable to such frauds since they can be easily fooled with common printed facial photographs. State-of-the-art approaches, based on Convolutional Neural Networks (CNNs), present good results in face spoofing detection. However, these methods do not consider the importance of learning deep local features from each facial region, even though it is known from face recognition that each facial region presents different visual aspects, which can also be exploited for face spoofing detection. In this work we propose a novel CNN architecture trained in two steps for such task. Initially, each part of the neural network learns features from a given facial region. Afterwards, the whole model is fine-tuned on the whole facial images. Results show that such pre-training step allows the CNN to learn different local spoofing cues, improving the performance and the convergence speed of the final model, outperforming the state-of-the-art approaches.