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

I’m a Ph.D. from University of California, San Diego (USA), in Electrical and Computer Engineering. Currently a Research Scientist at INESCTEC, Porto, and an invited professor at the Computer Science Department in the School of Engineering, University of Porto (Portugal).

My professional activities started in October 2000. Having received my Licenciatura† in Computer Science and Engineering from the University of Porto, I’ve started my career as a network engineer at Vodafone. There I was responsible for the design and operation of the core IP network, serving both internal and (external) corporate customers. In 2005, I’ve joined the IP division of Alcatel-Lucent in Lisbon, where my role was to design and customize network solutions for major telecom carriers in the EMEA region; performing proof-of-concept trials with potential customers. Concurrently to these professional activities, in 2003 I’ve completed an M.A. in Applied Mathematics under the supervision of Prof. Andre Puga, also at University of Porto. In 2008 I’ve joined the doctoral program on “Intelligent Systems, Robotics and Control” at U.C. San Diego, where I’ve concluded both M.Sc. (2011) and Ph.D. (2015) under the supervision of Prof. Nuno Vasconcelos. During my studies at UCSD I served as Research and Teaching Assistant; and I was lucky enough to do a brief internship at Telefonica R&D in Barcelona (Spain).

Interest
Topics
Details

Details

001
Publications

2018

Elastic deformations for data augmentation in breast cancer mass detection

Authors
Castro, E; Cardoso, JS; Pereira, JC;

Publication
2018 IEEE EMBS International Conference on Biomedical & Health Informatics, BHI 2018, Las Vegas, NV, USA, March 4-7, 2018

Abstract

2017

Digital Mammography DREAM Challenge: Participant Experience 2 (Conference Presentation)

Authors
Pereira, JC;

Publication
Medical Imaging 2017: Computer-Aided Diagnosis, Orlando, Florida, United States, 11-16 February 2017

Abstract

2016

Large Margin Discriminant Dimensionality Reduction in Prediction Space

Authors
Saberian, MohammadJ.; Pereira, JoseCosta; Vasconcelos, Nuno; Xu, Can;

Publication
Advances in Neural Information Processing Systems 29: Annual Conference on Neural Information Processing Systems 2016, December 5-10, 2016, Barcelona, Spain

Abstract

2015

Adaptation of Visual Models with Cross-modal Regularization

Authors
Costa Pereira, JMC;

Publication
base-search.net (ftcdlib:qt1bd3r86q)

Abstract

2014

On the Role of Correlation and Abstraction in Cross-Modal Multimedia Retrieval

Authors
Costa Pereira, JC; Coviello, E; Doyle, G; Rasiwasia, N; Lanckriet, GRG; Levy, R; Vasconcelos, N;

Publication
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE

Abstract
The problem of cross-modal retrieval from multimedia repositories is considered. This problem addresses the design of retrieval systems that support queries across content modalities, for example, using an image to search for texts. A mathematical formulation is proposed, equating the design of cross-modal retrieval systems to that of isomorphic feature spaces for different content modalities. Two hypotheses are then investigated regarding the fundamental attributes of these spaces. The first is that low-level cross-modal correlations should be accounted for. The second is that the space should enable semantic abstraction. Three new solutions to the cross-modal retrieval problem are then derived from these hypotheses: correlation matching (CM), an unsupervised method which models cross-modal correlations, semantic matching (SM), a supervised technique that relies on semantic representation, and semantic correlation matching (SCM), which combines both. An extensive evaluation of retrieval performance is conducted to test the validity of the hypotheses. All approaches are shown successful for text retrieval in response to image queries and vice versa. It is concluded that both hypotheses hold, in a complementary form, although evidence in favor of the abstraction hypothesis is stronger than that for correlation.