Victor Hugo Cunha de Melo
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Victor Hugo Cunha de Melo has a total of 29 co-authors. Here are the top co-authors by number of shared publications:
- William Robson Schwartz
Federal University of Minas Gerais
5 shared publications - Carlos Caetano
Instituto de Computação, Universidade Estadual de Campinas (UNICAMP)
2 shared publications - Igor Leonardo Oliveira Bastos
IFood Brasil, Osasco, Brazil
2 shared publications - Jefersson Alex dos Santos
University of Sheffield
2 shared publications - Ana Lúcia Brunialti Godard
Laboratório de Genética Animal e Humana, Departamento de Genética, Ecologia e Evolução, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
1 shared publication - Angelica Thomaz Vieira
UPR 9022 CNRS, IBMC, University of Strasbourg
1 shared publication
Latest Publications
- 17 May 2022
- Published by Springer Nature
- p. 429-441
DASP: Dual-autoencoder Architecture for Skin Prediction
- 17 May 2022
- Published by Springer Nature
- p. 429-441
DASP: Dual-autoencoder Architecture for Skin Prediction
- Open Access
- 22 February 2021
- Published by Public Library of Science (PLoS) in PLOS Neglected Tropical Diseases
- Vol. 15 (2) , e0009171
NLRP6-associated host microbiota composition impacts in the intestinal barrier to systemic dissemination of Brucella abortus
Brucella abortus is a Gram-negative bacterium responsible for a worldwide zoonotic infection—Brucellosis, which has been associated with high morbidity rate in humans and severe economic losses in infected livestock. The natural route of infection is through oral and nasal mucosa but the invasion process through host gut mucosa is yet to be understood. Studies have examined the role of NLRP6 (NOD-like receptor family pyrin domain-containing-6 protein) in gut homeostasis and defense against pathogens. Here, we investigated the impact of gut microbiota and NLRP6 in a murine model of Ba oral infection. Nlrp6-/- and wild-type (WT) mice were infected by oral gavage with Ba and tissues samples were collected at different time points. Our results suggest that Ba oral infection leads to significant alterations in gut microbiota. Moreover, Nlrp6-/- mice were more resistant to infection, with decreased CFU in the liver and reduction in gut permeability when compared to the control group. Fecal microbiota transplantation from WT and Nlrp6-/- into germ-free mice reflected the gut permeability phenotype from the donors. Additionally, depletion of gut microbiota by broad-spectrum-antibiotic treatment prevented Ba replication in WT while favoring bacterial growth in Nlrp6-/-. Finally, we observed higher eosinophils in the gut and leukocytes in the blood of infected Nlrp6-/- compared to WT-infected mice, which might be associated to the Nlrp6-/- resistance phenotype. Altogether, these results indicated that gut microbiota composition is the major factor involved in the initial stages of pathogen host replication and partially also by the resistance phenotype observed in Nlrp6 -/- mice regulating host inflammation against Ba infection.
- 22 February 2021
- Published by Public Library of Science (PLoS) in PLOS Neglected Tropical Diseases
- Vol. 15 (2) , e0009171
Open AccessNLRP6-associated host microbiota composition impacts in the intestinal barrier to systemic dissemination of Brucella abortus
Brucella abortus is a Gram-negative bacterium responsible for a worldwide zoonotic infection—Brucellosis, which has been associated with high morbidity rate in humans and severe economic losses in infected livestock. The natural route of infection is through oral and nasal mucosa but the invasion process through host gut mucosa is yet to be understood. Studies have examined the role of NLRP6 (NOD-like receptor family pyrin domain-containing-6 protein) in gut homeostasis and defense against pathogens. Here, we investigated the impact of gut microbiota and NLRP6 in a murine model of Ba oral infection. Nlrp6-/- and wild-type (WT) mice were infected by oral gavage with Ba and tissues samples were collected at different time points. Our results suggest that Ba oral infection leads to significant alterations in gut microbiota. Moreover, Nlrp6-/- mice were more resistant to infection, with decreased CFU in the liver and reduction in gut permeability when compared to the control group. Fecal microbiota transplantation from WT and Nlrp6-/- into germ-free mice reflected the gut permeability phenotype from the donors. Additionally, depletion of gut microbiota by broad-spectrum-antibiotic treatment prevented Ba replication in WT while favoring bacterial growth in Nlrp6-/-. Finally, we observed higher eosinophils in the gut and leukocytes in the blood of infected Nlrp6-/- compared to WT-infected mice, which might be associated to the Nlrp6-/- resistance phenotype. Altogether, these results indicated that gut microbiota composition is the major factor involved in the initial stages of pathogen host replication and partially also by the resistance phenotype observed in Nlrp6 -/- mice regulating host inflammation against Ba infection.
- Open Access
Covariance-free Partial Least Squares: An Incremental Dimensionality Reduction Method
Dimensionality reduction plays an important role in computer vision problems since it reduces computational cost and is often capable of yielding more discriminative data representation. In this context, Partial Least Squares (PLS) has presented notable results in tasks such as image classification and neural network optimization. However, PLS is infeasible on large datasets, such as ImageNet, because it requires all the data to be in memory in advance, which is often impractical due to hardware limitations. Additionally, this requirement prevents us from employing PLS on streaming applications where the data are being continuously generated. Motivated by this, we propose a novel incremental PLS, named Covariance-free Incremental Partial Least Squares (CIPLS), which learns a low-dimensional representation of the data using a single sample at a time. In contrast to other state-of-the-art approaches, instead of adopting a partially-discriminative or SGD-based model, we extend Nonlinear Iterative Partial Least Squares (NIPALS) -- the standard algorithm used to compute PLS -- for incremental processing. Among the advantages of this approach are the preservation of discriminative information across all components, the possibility of employing its score matrices for feature selection, and its computational efficiency. We validate CIPLS on face verification and image classification tasks, where it outperforms several other incremental dimensionality reduction techniques. In the context of feature selection, CIPLS achieves comparable results when compared to state-of-the-art techniques.
Open AccessCovariance-free Partial Least Squares: An Incremental Dimensionality Reduction Method
Dimensionality reduction plays an important role in computer vision problems since it reduces computational cost and is often capable of yielding more discriminative data representation. In this context, Partial Least Squares (PLS) has presented notable results in tasks such as image classification and neural network optimization. However, PLS is infeasible on large datasets, such as ImageNet, because it requires all the data to be in memory in advance, which is often impractical due to hardware limitations. Additionally, this requirement prevents us from employing PLS on streaming applications where the data are being continuously generated. Motivated by this, we propose a novel incremental PLS, named Covariance-free Incremental Partial Least Squares (CIPLS), which learns a low-dimensional representation of the data using a single sample at a time. In contrast to other state-of-the-art approaches, instead of adopting a partially-discriminative or SGD-based model, we extend Nonlinear Iterative Partial Least Squares (NIPALS) -- the standard algorithm used to compute PLS -- for incremental processing. Among the advantages of this approach are the preservation of discriminative information across all components, the possibility of employing its score matrices for feature selection, and its computational efficiency. We validate CIPLS on face verification and image classification tasks, where it outperforms several other incremental dimensionality reduction techniques. In the context of feature selection, CIPLS achieves comparable results when compared to state-of-the-art techniques.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
MORA: A Generative Approach to Extract Spatiotemporal Information Applied to Gesture Recognition
Gestures are related to a non-verbal language used on the interaction between subjects. Due to its applicability in several contexts, gesture recognition has been investigated by different researches, often investing on the capture of motion and appearance on videos. However, most of these methods do not properly explore the well-defined gesture temporal structure and are not suitable to deal with an increasing number of classes. Thus, we propose the Multi-Output Recurrent Autoencoders (MORA), an approach that relies on the representation of each gesture class independently. MORA employs a specific autoencoder model per class, composed by convolutional (3D) and a Gated Recurrent Unit (GRU) layer, what allows spatiotemporal information extraction and scalability in terms of number of classes. To validate MORA, experiments are conducted on SKIG and ChaLearn IsoGD datasets, for which the approach achieved accuracies comparable to state-of-the-art methods.
- 1 November 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
MORA: A Generative Approach to Extract Spatiotemporal Information Applied to Gesture Recognition
Gestures are related to a non-verbal language used on the interaction between subjects. Due to its applicability in several contexts, gesture recognition has been investigated by different researches, often investing on the capture of motion and appearance on videos. However, most of these methods do not properly explore the well-defined gesture temporal structure and are not suitable to deal with an increasing number of classes. Thus, we propose the Multi-Output Recurrent Autoencoders (MORA), an approach that relies on the representation of each gesture class independently. MORA employs a specific autoencoder model per class, composed by convolutional (3D) and a Gated Recurrent Unit (GRU) layer, what allows spatiotemporal information extraction and scalability in terms of number of classes. To validate MORA, experiments are conducted on SKIG and ChaLearn IsoGD datasets, for which the approach achieved accuracies comparable to state-of-the-art methods.
- Open Access
Activity Recognition based on a Magnitude-Orientation Stream Network
The temporal component of videos provides an important clue for activity recognition, as a number of activities can be reliably recognized based on the motion information. In view of that, this work proposes a novel temporal stream for two-stream convolutional networks based on images computed from the optical flow magnitude and orientation, named Magnitude-Orientation Stream (MOS), to learn the motion in a better and richer manner. Our method applies simple nonlinear transformations on the vertical and horizontal components of the optical flow to generate input images for the temporal stream. Experimental results, carried on two well-known datasets (HMDB51 and UCF101), demonstrate that using our proposed temporal stream as input to existing neural network architectures can improve their performance for activity recognition. Results demonstrate that our temporal stream provides complementary information able to improve the classical two-stream methods, indicating the suitability of our approach to be used as a temporal video representation.
Open AccessActivity Recognition based on a Magnitude-Orientation Stream Network
The temporal component of videos provides an important clue for activity recognition, as a number of activities can be reliably recognized based on the motion information. In view of that, this work proposes a novel temporal stream for two-stream convolutional networks based on images computed from the optical flow magnitude and orientation, named Magnitude-Orientation Stream (MOS), to learn the motion in a better and richer manner. Our method applies simple nonlinear transformations on the vertical and horizontal components of the optical flow to generate input images for the temporal stream. Experimental results, carried on two well-known datasets (HMDB51 and UCF101), demonstrate that using our proposed temporal stream as input to existing neural network architectures can improve their performance for activity recognition. Results demonstrate that our temporal stream provides complementary information able to improve the classical two-stream methods, indicating the suitability of our approach to be used as a temporal video representation.
- Open Access
- 26 August 2015
- Published by Sociedade Brasileira de Computação
- p. 21-30
A Computational Infrastructure Model for Research on Computer Vision
With the advances on science, a powerful computational infrastructure is desirable to increase the performance of experiments. The same holds true for research on the Computer Vision field, which deals with large amounts of data and also requires intensive computation. Nevertheless, the administration of a computational infrastructure involves many tasks, such as system configuration, preventive maintenance and storage management, which becomes very challenging for many research groups. With that in mind, this work proposes an infrastructure model to assist researchers with focus on Computer Vision. We conducted a series of tests to evaluate the performance of our model and the processing power of our infrastructure.
- 26 August 2015
- Published by Sociedade Brasileira de Computação
- p. 21-30
Open AccessA Computational Infrastructure Model for Research on Computer Vision
With the advances on science, a powerful computational infrastructure is desirable to increase the performance of experiments. The same holds true for research on the Computer Vision field, which deals with large amounts of data and also requires intensive computation. Nevertheless, the administration of a computational infrastructure involves many tasks, such as system configuration, preventive maintenance and storage management, which becomes very challenging for many research groups. With that in mind, this work proposes an infrastructure model to assist researchers with focus on Computer Vision. We conducted a series of tests to evaluate the performance of our model and the processing power of our infrastructure.