Cookies
Usamos cookies para melhorar nosso site e a sua experiência. Ao continuar a navegar no site, você aceita a nossa política de cookies. Ver mais
Aceitar Rejeitar
  • Menu
Sobre
Download foto HD

Sobre

Ana F. Sequeira é licenciada em Matemática, desde 2002, Mestre em Engenharia Matemática, desde 2007, pela Faculdade de Ciências e doutorada em Engenharia e Eletrotécnica e de Computadores, desde 2015, pela Faculdade de Engenharia, ambas as faculdades da Universidade do Porto.

Ana F. Sequeira colaborou com o INESC TEC como investigadora durante o seu doutoramento que visou as áreas de visão computacional e "machine learning" com foco em metodologias de detecção de vivacidade em íris e impressão digital.

Após a conclusão do doutoramento, Ana F. Sequeira colaborou na Universidade de Reading, UK, em dois projectos europeus relacionados com a aplicação de reconhecimento biométrico em controlo de fronteiras (FASTPASS e PROTECT).

A esta actividade seguiu-se uma colaboração a curto-prazo com a empresa Irisguard UK com o objectivo de pesquisar vulnerabilidades do produto EyePay® e desenvolver um protótipo de uma medida de protecção contra “spoofing attacks”.

Actualmente, Ana F. Sequeira colabora novamente com o INESC TEC como investigadora contratado.

Enquanto doutoranda e pós-doc, desde 2011, Ana F. Sequeira é coautora de vários artigos incluindo conferencias internacionais e revistas reconhecidas pela comunidade por citações; assim como liderou a criação de bases de dados e organização de eventos como competições e eventos.

Ao longo da sua  actividade de investigação Ana F. Sequeira adquiriu vasta experiência não apenas em tópicos de visão computacional/processamento de imagem mas também na aplicação de técnicas diversificadas de “machine learning”, desde as metodologias clássicas até as de “deep learning”.

Tópicos
de interesse
Detalhes

Detalhes

001
Publicações

2021

Maximum Relevance Minimum Redundancy Dropout with Informative Kernel Determinantal Point Process

Autores
Saffari, M; Khodayar, M; Saadabadi, MSE; Sequeira, AF; Cardoso, JS;

Publicação
Sensors

Abstract
In recent years, deep neural networks have shown significant progress in computer vision due to their large generalization capacity; however, the overfitting problem ubiquitously threatens the learning process of these highly nonlinear architectures. Dropout is a recent solution to mitigate overfitting that has witnessed significant success in various classification applications. Recently, many efforts have been made to improve the Standard dropout using an unsupervised merit-based semantic selection of neurons in the latent space. However, these studies do not consider the task-relevant information quality and quantity and the diversity of the latent kernels. To solve the challenge of dropping less informative neurons in deep learning, we propose an efficient end-to-end dropout algorithm that selects the most informative neurons with the highest correlation with the target output considering the sparsity in its selection procedure. First, to promote activation diversity, we devise an approach to select the most diverse set of neurons by making use of determinantal point process (DPP) sampling. Furthermore, to incorporate task specificity into deep latent features, a mutual information (MI)-based merit function is developed. Leveraging the proposed MI with DPP sampling, we introduce the novel DPPMI dropout that adaptively adjusts the retention rate of neurons based on their contribution to the neural network task. Empirical studies on real-world classification benchmarks including, MNIST, SVHN, CIFAR10, CIFAR100, demonstrate the superiority of our proposed method over recent state-of-the-art dropout algorithms in the literature.

2020

Interpretable Biometrics: Should We Rethink How Presentation Attack Detection is Evaluated?

Autores
Sequeira, AF; Silva, W; Pinto, JR; Goncalves, T; Cardoso, JS;

Publicação
8th International Workshop on Biometrics and Forensics, IWBF 2020, Porto, Portugal, April 29-30, 2020

Abstract
Presentation attack detection (PAD) methods are commonly evaluated using metrics based on the predicted labels. This is a limitation, especially for more elusive methods based on deep learning which can freely learn the most suitable features. Though often being more accurate, these models operate as complex black boxes which makes the inner processes that sustain their predictions still baffling. Interpretability tools are now being used to delve deeper into the operation of machine learning methods, especially artificial networks, to better understand how they reach their decisions. In this paper, we make a case for the integration of interpretability tools in the evaluation of PAD. A simple model for face PAD, based on convolutional neural networks, was implemented and evaluated using both traditional metrics (APCER, BPCER and EER) and interpretability tools (Grad-CAM), using data from the ROSE Youtu video collection. The results show that interpretability tools can capture more completely the intricate behavior of the implemented model, and enable the identification of certain properties that should be verified by a PAD method that is robust, coherent, meaningful, and can adequately generalize to unseen data and attacks. One can conclude that, with further efforts devoted towards higher objectivity in interpretability, this can be the key to obtain deeper and more thorough PAD performance evaluation setups. © 2020 IEEE.

2020

A robust fingerprint presentation attack detection method against unseen attacks through adversarial learning

Autores
Afonso Pereira, J; Sequeira, AF; Pernes, D; Cardoso, JS;

Publicação
BIOSIG 2020 - Proceedings of the 19th International Conference of the Biometrics Special Interest Group

Abstract
Fingerprint presentation attack detection (PAD) methods present a stunning performance in current literature. However, the fingerprint PAD generalisation problem is still an open challenge requiring the development of methods able to cope with sophisticated and unseen attacks as our eventual intruders become more capable. This work addresses this problem by applying a regularisation technique based on an adversarial training and representation learning specifically designed to to improve the PAD generalisation capacity of the model to an unseen attack. In the adopted approach, the model jointly learns the representation and the classifier from the data, while explicitly imposing invariance in the high-level representations regarding the type of attacks for a robust PAD. The application of the adversarial training methodology is evaluated in two different scenarios: i) a handcrafted feature extraction method combined with a Multilayer Perceptron (MLP); and ii) an end-to-end solution using a Convolutional Neural Network (CNN). The experimental results demonstrated that the adopted regularisation strategies equipped the neural networks with increased PAD robustness. The adversarial approach particularly improved the CNN models' capacity for attacks detection in the unseen-attack scenario, showing remarkable improved APCER error rates when compared to state-of-the-art methods in similar conditions. © 2020 German Computer Association (Gesellschaft für Informatik e.V.).

2020

PROTECT: Pervasive and useR fOcused biomeTrics bordEr projeCT - a case study

Autores
Galdi, C; Boyle, J; Chen, LL; Chiesa, V; Debiasi, L; Dugelay, JL; Ferryman, J; Grudzien, A; Kauba, C; Kirchgasser, S; Kowalski, M; Linortner, M; Maik, P; Michon, K; Patino, L; Prommegger, B; Sequeira, AF; Szklarski, L; Uhl, A;

Publicação
IET BIOMETRICS

Abstract
Pervasive and useR fOcused biomeTrics bordEr projeCT (PROTECT) is an EU project funded by the Horizon 2020 research and Innovation Programme. The main aim of PROTECT was to build an advanced biometric-based person identification system that works robustly across a range of border crossing types and that has strong user-centric features. This work presents the case study of the multibiometric verification system developed within PROTECT. The system has been developed to be suitable for different borders such as air, sea, and land borders. The system covers two use cases: the walk-through scenario, in which the traveller is on foot; the drive-through scenario, in which the traveller is in a vehicle. Each deployment includes a different set of biometric traits and this study illustrates how to evaluate such multibiometric system in accordance with international standards and, in particular, how to overcome practical problems that may be encountered when dealing with multibiometric evaluation, such as different score distributions and missing scores.

2020

Multimedia systems and applications in biomedicine

Autores
Domingues, I; Sequeira, AF; Pinto, C; Rocha,;

Publicação
COMPUTER METHODS IN BIOMECHANICS AND BIOMEDICAL ENGINEERING-IMAGING AND VISUALIZATION

Abstract

Teses
supervisionadas

2020

Head Pose Estimation for Biometric Recognition Systems

Autor
João Manuel Guedes Ferreira

Instituição
UP-FEUP

2020

Explainable Artificial Intelligence For Biometric Analysis

Autor
Pedro Carneiro Neto

Instituição
UP-FEUP

2020

Face biOmetrics UNder severe representation Drifts

Autor
Mohsen Saffari

Instituição
UP-FEUP

2020

Fingerprint Anti Spoofing – Domain Adaptation and Adversarial Learning

Autor
João Afonso Pinto Pereira

Instituição
UP-FEUP