A
Alberto F. De Souza
Researcher at Universidade Federal do Espírito Santo
Publications - 125
Citations - 3118
Alberto F. De Souza is an academic researcher from Universidade Federal do Espírito Santo. The author has contributed to research in topics: Artificial neural network & Deep learning. The author has an hindex of 19, co-authored 119 publications receiving 1923 citations. Previous affiliations of Alberto F. De Souza include Federal University of Rio de Janeiro & University College London.
Papers
More filters
Journal ArticleDOI
Facial expression recognition with Convolutional Neural Networks
TL;DR: A simple solution for facial expression recognition that uses a combination of Convolutional Neural Network and specific image pre-processing steps to extract only expression specific features from a face image and explore the presentation order of the samples during training.
Journal ArticleDOI
Self-driving cars: A survey
Claudine Badue,Rânik Guidolini,Raphael V. Carneiro,Pedro Azevedo,Vinicius B. Cardoso,Avelino Forechi,Luan F. R. Jesus,Rodrigo F. Berriel,Thiago M. Paixão,Filipe Mutz,Lucas de Paula Veronese,Thiago Oliveira-Santos,Alberto F. De Souza +12 more
TL;DR: A detailed description of the architecture of the autonomy system of the self-driving car developed at the Universidade Federal do Espirito Santo (UFES), named Intelligent Autonomous Robotics Automobile (IARA), is presented.
Journal ArticleDOI
A Dataset for Improved RGBD-Based Object Detection and Pose Estimation for Warehouse Pick-and-Place
TL;DR: This letter provides a new rich dataset for advancing the state-of-the-art in RGBD-based 3D object pose estimation, which is focused on the challenges that arise when solving warehouse pick-and-place tasks.
Journal ArticleDOI
Prediction-based portfolio optimization model using neural networks
TL;DR: This work used neural network predictors to predict stocks' returns and derived a risk measure, based on the prediction errors, that have the same statistical foundation of the mean-variance model, and showed that it is possible to obtain Normal prediction errors with non-Normal time series of stock returns.
Proceedings ArticleDOI
Copycat CNN: Stealing Knowledge by Persuading Confession with Random Non-Labeled Data
Jacson Rodrigues Correia-Silva,Rodrigo F. Berriel,Claudine Badue,Alberto F. De Souza,Thiago Oliveira-Santos +4 more
TL;DR: This paper investigates if a target blackbox CNN can be copied by persuading it to confess its knowledge through random non-labeled data, and shows that it is possible to create a copycat CNN by simply querying a target network as black-box withrandom non- labeled data.