Cookies Policy
We use cookies to improve our site and your experience. By continuing to browse our site you accept our cookie policy. Find out More
Close
  • Menu
About

About

Ana Rebelo was born in Porto, Portugal, in 1985. She graduated in Mathematics Applied to Technology at School of Sciences of University of Porto, Portugal, in 2007. She received the M.Sc. degree in Mathematical Enginnering from School of Sciences of University of Porto, Portugal, in 2008. In 2008, she started her Ph.D. studies at School of Engineering, University of Porto, Portugal. She has been working since 2007 as a researcher at INESC TEC, an R&D institute affiliated to University of Porto, in the Visual Computing and Machine Intelligence Group (VCMI). Ana Rebelo was a Project Member of one FCT (Foundation of Science and Technology - Portugal) research project in the area of optical music recognition (OMR). She is currently working at Escola Superior Biotecnologia, Universidade Católica Portuguesa as an Invited Assistant Professor and she is a FCT PostDoctoral Researcher at INESC TEC. Her main research interests include computer vision, image processing, biometrics and document analysis.

Interest
Topics
Details

Details

004
Publications

2019

On the role of multimodal learning in the recognition of sign language

Authors
Ferreira, PM; Cardoso, JS; Rebelo, A;

Publication
Multimedia Tools and Applications

Abstract
Sign Language Recognition (SLR) has become one of the most important research areas in the field of human computer interaction. SLR systems are meant to automatically translate sign language into text or speech, in order to reduce the communicational gap between deaf and hearing people. The aim of this paper is to exploit multimodal learning techniques for an accurate SLR, making use of data provided by Kinect and Leap Motion. In this regard, single-modality approaches as well as different multimodal methods, mainly based on convolutional neural networks, are proposed. Our main contribution is a novel multimodal end-to-end neural network that explicitly models private feature representations that are specific to each modality and shared feature representations that are similar between modalities. By imposing such regularization in the learning process, the underlying idea is to increase the discriminative ability of the learned features and, hence, improve the generalization capability of the model. Experimental results demonstrate that multimodal learning yields an overall improvement in the sign recognition performance. In particular, the novel neural network architecture outperforms the current state-of-the-art methods for the SLR task. © 2018, Springer Science+Business Media, LLC, part of Springer Nature.

2018

Physiological Inspired Deep Neural Networks for Emotion Recognition

Authors
Ferreira, PM; Marques, F; Cardoso, JS; Rebelo, A;

Publication
IEEE ACCESS

Abstract
Facial expression recognition (FER) is currently one of the most active research topics due to its wide range of applications in the human-computer interaction field. An important part of the recent success of automatic FER was achieved thanks to the emergence of deep learning approaches. However, training deep networks for FER is still a very challenging task, since most of the available FER data sets are relatively small. Although transfer learning can partially alleviate the issue, the performance of deep models is still below of its full potential as deep features may contain redundant information from the pre-trained domain. Instead, we propose a novel end-to-end neural network architecture along with a well-designed loss function based on the strong prior knowledge that facial expressions are the result of the motions of some facial muscles and components. The loss function is defined to regularize the entire learning process so that the proposed neural network is able to explicitly learn expression-specific features. Experimental results demonstrate the effectiveness of the proposed model in both lab-controlled and wild environments. In particular, the proposed neural network provides quite promising results, outperforming in most cases the current state-of-the-art methods.

2018

Are Deep Learning Methods Ready for Prime Time in Fingerprints Minutiae Extraction?

Authors
Rebelo, A; Oliveira, T; Correia, ME; Cardoso, JS;

Publication
Progress in Pattern Recognition, Image Analysis, Computer Vision, and Applications - Lecture Notes in Computer Science

Abstract

2017

Multimodal Learning for Sign Language Recognition

Authors
Ferreira, PM; Cardoso, JS; Rebelo, A;

Publication
Pattern Recognition and Image Analysis - Lecture Notes in Computer Science

Abstract

2016

A Directed Acyclic Graph-Large Margin Distribution Machine Model for Music Symbol Classification

Authors
Wen, CH; Zhang, J; Rebelo, A; Cheng, FY;

Publication
PLOS ONE

Abstract
Optical Music Recognition (OMR) has received increasing attention in recent years. In this paper, we propose a classifier based on a new method named Directed Acyclic Graph-Large margin Distribution Machine (DAG-LDM). The DAG-LDM is an improvement of the Large margin Distribution Machine (LDM), which is a binary classifier that optimizes the margin distribution by maximizing the margin mean and minimizing the margin variance simultaneously. We modify the LDM to the DAG-LDM to solve the multi-class music symbol classification problem. Tests are conducted on more than 10000 music symbol images, obtained from handwritten and printed images of music scores. The proposed method provides superior classification capability and achieves much higher classification accuracy than the state-of-the-art algorithms such as Support Vector Machines (SVMs) and Neural Networks (NNs).

Supervised
thesis

2015

Smart Image Retargeting Techniques

Author
Laura Figueiredo Ângelo

Institution
UP-FEUP

2015

Portuguese Sign Language recognition

Author
Pedro Miguel Martins Ferreira

Institution
UP-FEUP