no code implementations • 16 Mar 2024 • Mahdi Alehdaghi, Pourya Shamsolmoali, Rafael M. O. Cruz, Eric Granger
In particular, our method minimizes the cross-modal gap by identifying and aligning shared prototypes that capture key discriminative features across modalities, then uses multiple bridging steps based on this information to enhance the feature representation.
no code implementations • 20 Dec 2023 • Abdallah Dib, Luiz Gustavo Hafemann, Emeline Got, Trevor Anderson, Amin Fadaeinejad, Rafael M. O. Cruz, Marc-Andre Carbonneau
We also introduce a new dataset, named FFHQ-UV-Intrinsics, the first public dataset providing intrinsic face attributes at scale (diffuse, specular, ambient occlusion and translucency maps) for a total of 10k subjects.
Ranked #4 on 3D Face Reconstruction on REALY
1 code implementation • 25 Sep 2023 • Paulo R. G. Cordeiro, George D. C. Cavalcanti, Rafael M. O. Cruz
To evaluate this idea, we introduce the Post-Selection Dynamic Ensemble Selection (PS-DES) approach, a post-selection scheme that evaluates ensembles selected by several DES techniques using different metrics.
no code implementations • 6 Jul 2023 • Mahdi Alehdaghi, Arthur Josi, Pourya Shamsolmoali, Rafael M. O. Cruz, Eric Granger
In this paper, the Adaptive Generation of Privileged Intermediate Information training approach is introduced to adapt and generate a virtual domain that bridges discriminant information between the V and I modalities.
no code implementations • 30 Jun 2023 • Eduardo V. L. Barboza, Paulo R. Lisboa de Almeida, Alceu de Souza Britto Jr, Rafael M. O. Cruz
Data normalization is an essential task when modeling a classification system.
1 code implementation • 29 Apr 2023 • Arthur Josi, Mahdi Alehdaghi, Rafael M. O. Cruz, Eric Granger
For realistic evaluation of multimodal (and cross-modal) V-I person ReID models, we propose new challenging corrupted datasets for scenarios where V and I cameras are co-located (CL) and not co-located (NCL).
1 code implementation • 23 Dec 2022 • Lucas B. V. de Amorim, George D. C. Cavalcanti, Rafael M. O. Cruz
In this paper, we execute a broad experiment comparing the impact of 5 scaling techniques on the performances of 20 classification algorithms among monolithic and ensemble models, applying them to 82 publicly available datasets with varying imbalance ratios.
1 code implementation • 22 Nov 2022 • Arthur Josi, Mahdi Alehdaghi, Rafael M. O. Cruz, Eric Granger
Several deep learning models have been proposed for visible-infrared (V-I) person ReID to recognize individuals from images captured using RGB and IR cameras.
1 code implementation • 19 Sep 2022 • Mahdi Alehdaghi, Arthur Josi, Rafael M. O. Cruz, Eric Granger
% This paper introduces a novel approach for a creating intermediate virtual domain that acts as bridges between the two main domains (i. e., RGB and IR modalities) during training.
1 code implementation • 16 Jun 2022 • Mariana A. Souza, Robert Sabourin, George D. C. Cavalcanti, Rafael M. O. Cruz
Class imbalance is a characteristic known for making learning more challenging for classification models as they may end up biased towards the majority class.
2 code implementations • 20 May 2022 • Reza Davtalab, Rafael M. O. Cruz, Robert Sabourin
Most dynamic ensemble selection (DES) methods utilize the K-Nearest Neighbors (KNN) algorithm to estimate the competence of classifiers in a small region surrounding the query sample.
1 code implementation • 12 May 2022 • Félix Remigereau, Djebril Mekhazni, Sajjad Abdoli, Le Thanh Nguyen-Meidine, Rafael M. O. Cruz, Eric Granger
Despite the recent success of deep learning architectures, person re-identification (ReID) remains a challenging problem in real-word applications.
1 code implementation • 18 Jan 2022 • Rafael M. O. Cruz, Woshington V. de Sousa, George D. C. Cavalcanti
This work argues that a combination of multiple feature extraction techniques and different classification models is needed.
no code implementations • 19 Oct 2020 • Victor L. F. Souza, Adriano L. I. Oliveira, Rafael M. O. Cruz, Robert Sabourin
We proposed a method based on a global validation strategy with an external archive to control overfitting during the search for the most discriminant representation.
no code implementations • 7 Apr 2020 • Victor L. F. Souza, Adriano L. I. Oliveira, Rafael M. O. Cruz, Robert Sabourin
This paper investigates the presence of overfitting when using Binary Particle Swarm Optimization (BPSO) to perform the feature selection in a context of Handwritten Signature Verification (HSV).
no code implementations • 3 Apr 2020 • Victor L. F. Souza, Adriano L. I. Oliveira, Rafael M. O. Cruz, Robert Sabourin
Among the advantages of this framework is its scalability to deal with some of these challenges and its ease in managing new writers, and hence of being used in a transfer learning context.
1 code implementation • 1 Apr 2020 • Mariana A. Souza, Robert Sabourin, George D. C. Cavalcanti, Rafael M. O. Cruz
Our proposed framework builds a multi-label meta-classifier responsible for recommending a set of relevant model types based on the local data complexity of the region surrounding each test sample.
no code implementations • 22 Nov 2018 • Rafael M. O. Cruz, Mariana A. Souza, Robert Sabourin, George D. C. Cavalcanti
Hence, this paper presents an empirical analysis of dynamic selection techniques and data preprocessing methods for dealing with multi-class imbalanced problems.
no code implementations • 1 Nov 2018 • Rafael M. O. Cruz, Robert Sabourin, George D. C. Cavalcanti
The key issue in Dynamic Ensemble Selection (DES) is defining a suitable criterion for calculating the classifiers' competence.
no code implementations • 1 Nov 2018 • Rafael M. O. Cruz, Robert Sabourin, George D. C. Cavalcanti
The meta-features are computed using the training data and used to train a meta-classifier that is able to predict whether or not a base classifier from the pool is competent enough to classify an input instance.
no code implementations • 1 Nov 2018 • Rafael M. O. Cruz, Robert Sabourin, George D. C. Cavalcanti
In this paper, we propose improvements to the training and generalization phase of the META-DES framework.
no code implementations • 1 Nov 2018 • Rafael M. O. Cruz, Robert Sabourin, George D. C. Cavalcanti
The more important step in DES techniques is estimating the competence of the base classifiers for the classification of each specific test sample.
no code implementations • 1 Nov 2018 • Rafael M. O. Cruz, George D. C. Cavalcanti, Tsang Ing Ren
Dynamic classifier selection systems aim to select a group of classifiers that is most adequate for a specific query pattern.
no code implementations • 1 Oct 2018 • Rafael M. O. Cruz, Dayvid V. R. Oliveira, George D. C. Cavalcanti, Robert Sabourin
Despite being very effective in several classification tasks, Dynamic Ensemble Selection (DES) techniques can select classifiers that classify all samples in the region of competence as being from the same class.
no code implementations • 30 Sep 2018 • Rafael M. O. Cruz, Robert Sabourin, George D. C. Cavalcanti, Tsang Ing Ren
The meta-features are extracted from the training data and used to train a meta-classifier to predict whether or not a base classifier is competent enough to classify an input instance.
no code implementations • 5 Sep 2018 • Mariana A. Souza, George D. C. Cavalcanti, Rafael M. O. Cruz, Robert Sabourin
Thus, we propose in this work an online pool generation method that produces a locally accurate pool for test samples in difficult regions of the feature space.
no code implementations • 21 Apr 2018 • Rafael M. O. Cruz, Hiba H. Zakane, Robert Sabourin, George D. C. Cavalcanti
Experiments are performed on 18 state-of-the-art DS techniques over 30 classification datasets and results show that DS methods present a significant boost in classification accuracy even though they use the same neighborhood as the K-NN.
no code implementations • 20 Apr 2018 • Felipe N. Walmsley, George D. C. Cavalcanti, Dayvid V. R. Oliveira, Rafael M. O. Cruz, Robert Sabourin
Techniques such as Bagging and Boosting have been successfully applied to a variety of problems.
no code implementations • 18 Apr 2018 • Dayvid V. R. Oliveira, George D. C. Cavalcanti, Thyago N. Porpino, Rafael M. O. Cruz, Robert Sabourin
The K-Nearest Oracles Eliminate (KNORA-E) DES selects all classifiers that correctly classify all samples in the region of competence of the test sample, if such classifier exists, otherwise, it removes from the region of competence the sample that is furthest from the test sample, and the process repeats.
no code implementations • 11 Mar 2018 • Rafael M. O. Cruz, Robert Sabourin, George D. C. Cavalcanti
Hence, this paper presents an empirical analysis of dynamic selection techniques and data preprocessing methods for dealing with multi-class imbalanced problems.
2 code implementations • 14 Feb 2018 • Rafael M. O. Cruz, Luiz G. Hafemann, Robert Sabourin, George D. C. Cavalcanti
DESlib is an open-source python library providing the implementation of several dynamic selection techniques.
no code implementations • 2 Sep 2015 • Rafael M. O. Cruz, Robert Sabourin, George D. C. Cavalcanti
In order to perform a more robust ensemble selection, we proposed the META-DES framework using meta-learning, where multiple criteria are encoded as meta-features and are passed down to a meta-classifier that is trained to estimate the competence level of a given classifier.