Publications

Publications by Ana Filipa Sequeira

2021

Chairs’ Message - 20<sup>th</sup> anniversary of BIOSIG

Authors
Brömme A.; Busch C.; Damer N.; Dantcheva A.; Gomez-Barrero M.; Raja K.; Rathgeb C.; Sequeira A.F.; Uhl A.;

Publication
Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)

Abstract

2021

Chairs' Message - 20th anniversary of BIOSIG

Authors
Brömme A.; Busch C.; Damer N.; Dantcheva A.; Gomez-Barrero M.; Raja K.; Rathgeb C.; Sequeira A.F.; Uhl A.;

Publication
BIOSIG 2021 - Proceedings of the 20th International Conference of the Biometrics Special Interest Group

Abstract

2021

Selected Works from International Workshop on Biometrics and Forensics 2020

Authors
Sequeira, AF; Ross, A;

Publication
IEEE Transactions on Biometrics, Behavior, and Identity Science

Abstract

2022

Chairs' Message

Authors
Brömme A.; Damer N.; Gomez-Barrero M.; Raja K.; Rathgeb C.; Sequeira A.F.; Todisco M.; Uhl A.;

Publication
BIOSIG 2022 - Proceedings of the 21st International Conference of the Biometrics Special Interest Group

Abstract

2023

Unveiling the Two-Faced Truth: Disentangling Morphed Identities for Face Morphing Detection

Authors
Caldeira, E; Neto, PC; Gonçalves, T; Damer, N; Sequeira, AF; Cardoso, JS;

Publication
31st European Signal Processing Conference, EUSIPCO 2023, Helsinki, Finland, September 4-8, 2023

Abstract
Morphing attacks keep threatening biometric systems, especially face recognition systems. Over time they have become simpler to perform and more realistic, as such, the usage of deep learning systems to detect these attacks has grown. At the same time, there is a constant concern regarding the lack of interpretability of deep learning models. Balancing performance and interpretability has been a difficult task for scientists. However, by leveraging domain information and proving some constraints, we have been able to develop IDistill, an interpretable method with state-of-the-art performance that provides information on both the identity separation on morph samples and their contribution to the final prediction. The domain information is learnt by an autoencoder and distilled to a classifier system in order to teach it to separate identity information. When compared to other methods in the literature it outperforms them in three out of five databases and is competitive in the remaining. © 2023 European Signal Processing Conference, EUSIPCO. All rights reserved.

CloseRead Abstract

2022

Explainable Biometrics in the Age of Deep Learning

Authors
Neto, PC; Gonçalves, T; Pinto, JR; Silva, W; Sequeira, AF; Ross, A; Cardoso, JS;

Publication
CoRR

Abstract