2 code implementations • 17 Feb 2023 • Nguyen Huu Phong, Bernardete Ribeiro
To extend our approach to video, we integrate ConvNets with state-of-the-art temporal methods such as Transformer and Recurrent Neural Networks.
Ranked #75 on Action Recognition on UCF101
1 code implementation • 20 May 2022 • Nguyen Huu Phong, Augusto Santos, Bernardete Ribeiro
In such framework, the vector of weights of each ConvNet is typically cast as the position of a particle in phase space whereby PSO collaborative dynamics intertwines with Stochastic Gradient Descent (SGD) in order to boost training performance and generalization.
Ranked #39 on Image Classification on CIFAR-10
no code implementations • 20 May 2022 • Nguyen Huu Phong, Bernardete Ribeiro
In this research, we present our findings to recognize American Sign Language from series of hand gestures.
1 code implementation • 30 Jul 2020 • Nguyen Huu Phong, Bernardete Ribeiro
In addition, we extend the training strategy so that our model performs comparably to leading models and can even match the state-of-the-art models on several challenging datasets (e. g., SVHN (0. 99), Cifar-100 (0. 9027) and Cifar-10 (0. 9852)).
Ranked #1 on Image Classification on Surrey ASL
no code implementations • 30 Jul 2020 • Nguyen Huu Phong, Bernardete Ribeiro
Capsule Networks face a critical problem in computer vision in the sense that the image background can challenge its performance, although they learn very well on training data.
1 code implementation • 1 Apr 2020 • Filipe Assunção, Nuno Lourenço, Bernardete Ribeiro, Penousal Machado
Despite aiding non-expert users to design and train ANNs, the vast majority of NE approaches disregard the knowledge that is gathered when solving other tasks, i. e., evolution starts from scratch for each problem, ultimately delaying the evolutionary process.
1 code implementation • 1 Apr 2020 • Filipe Assunção, Nuno Lourenço, Bernardete Ribeiro, Penousal Machado
The deployment of Machine Learning (ML) models is a difficult and time-consuming job that comprises a series of sequential and correlated tasks that go from the data pre-processing, and the design and extraction of features, to the choice of the ML algorithm and its parameterisation.
no code implementations • 8 May 2019 • Filipe Assunção, Nuno Lourenço, Penousal Machado, Bernardete Ribeiro
This paper proposes a new extension to Deep Evolutionary Network Structured Evolution (DENSER), called Fast-DENSER++ (F-DENSER++).
no code implementations • 18 Mar 2019 • Nguyen Huu Phong, Bernardete Ribeiro
Image recognition using Deep Learning has been evolved for decades though advances in the field through different settings is still a challenge.
no code implementations • 18 Mar 2019 • Nguyen Huu Phong, Bernardete Ribeiro
Capsule Networks (CN) offer new architectures for Deep Learning (DL) community.
no code implementations • 20 Dec 2018 • Filipe Rodrigues, Stanislav S. Borysov, Bernardete Ribeiro, Francisco C. Pereira
Public special events, like sports games, concerts and festivals are well known to create disruptions in transportation systems, often catching the operators by surprise.
no code implementations • 17 Aug 2018 • Filipe Rodrigues, Mariana Lourenço, Bernardete Ribeiro, Francisco Pereira
The growing need to analyze large collections of documents has led to great developments in topic modeling.
17 code implementations • 4 Jan 2018 • Filipe Assunção, Nuno Lourenço, Penousal Machado, Bernardete Ribeiro
Deep Evolutionary Network Structured Representation (DENSER) is a novel approach to automatically design Artificial Neural Networks (ANNs) using Evolutionary Computation.
no code implementations • 23 Sep 2017 • Jorge Rivero, Bernardete Ribeiro, Ning Chen, Fátima Silva Leite
This can be overcome with Zero-Shot Learning, a new approach in the field of Computer Vision, which can be described in two stages: the Attribute Learning and the Inference Stage.
no code implementations • 26 Jun 2017 • Filipe Assunção, Nuno Lourenço, Penousal Machado, Bernardete Ribeiro
On the other, there is no way to evolve networks with more than one output neuron.
no code implementations • 28 Jul 2016 • Jorge Luis Rivero Pérez, Bernardete Ribeiro
In this paper we propose a new algorithm for the attribute learning stage of ZSL.
no code implementations • 20 Apr 2016 • Gonçalo Oliveira, Xavier Frazão, André Pimentel, Bernardete Ribeiro
Brand recognition is a very challenging topic with many useful applications in localization recognition, advertisement and marketing.
no code implementations • 17 Apr 2016 • Jorge Luis Rivero Perez, Bernardete Ribeiro, Carlos Morell Perez
This approach hybridizes a Mahalanobis distance metric learning algorithm and a k-NN data stream classification algorithm with concept drift detection.