S
Samuel G. Fadel
Researcher at State University of Campinas
Publications - 17
Citations - 473
Samuel G. Fadel is an academic researcher from State University of Campinas. The author has contributed to research in topics: Artificial neural network & Dimensionality reduction. The author has an hindex of 6, co-authored 16 publications receiving 355 citations. Previous affiliations of Samuel G. Fadel include Lüneburg University & Spanish National Research Council.
Papers
More filters
Journal ArticleDOI
Visualizing the Hidden Activity of Artificial Neural Networks
TL;DR: It is shown how visualization can provide highly valuable feedback for network designers through experiments conducted in three traditional image classification benchmark datasets, and the presence of interpretable clusters of learned representations and the partitioning of artificial neurons into groups with apparently related discriminative roles are discovered.
Journal ArticleDOI
Exploiting ConvNet Diversity for Flooding Identification
Keiller Nogueira,Samuel G. Fadel,Icaro Cavalcante Dourado,Rafael de Oliveira Werneck,Javier A. V. Muñoz,Otavio A. B. Penatti,Rodrigo Tripodi Calumby,Lin Tzy Li,Jefersson A. dos Santos,Ricardo da Silva Torres +9 more
TL;DR: Several methods to perform flooding identification in high-resolution remote sensing images using deep learning are proposed, some based upon unique networks, such as dilated and deconvolutional ones, whereas others were conceived to exploit diversity of distinct networks in order to extract the maximum performance of each classifier.
Journal ArticleDOI
Nmap: A Novel Neighborhood Preservation Space-filling Algorithm
Felipe Simões Lage Gomes Duarte,Fabio Sikansi,Francisco M. Fatore,Samuel G. Fadel,Fernando V. Paulovich +4 more
TL;DR: A novel approach, called Neighborhood Treemap (Nmap), that seeks to solve the limitation of distance-similarity metaphor by employing a slice and scale strategy where the visual space is successively bisected on the horizontal or vertical directions and the bisections are scaled until one rectangle is defined per data element.
Journal ArticleDOI
LoCH: A neighborhood-based multidimensional projection technique for high-dimensional sparse spaces
TL;DR: A novel local technique specially designed for projecting heavy tail distance distributions, such as the one produced by high-dimensional sparse spaces, called Local Convex Hull (LoCH), relies on an iterative process that seeks to place each point close to the convex hull of its nearest neighbors.
Data-Driven flood detection using neural networks
Keiller Nogueira,Samuel G. Fadel,Icaro Cavalcante Dourado,Rafael de Oliveira Werneck,Javier A. V. Muñoz,Otavio A. B. Penatti,Rodrigo Tripodi Calumby,Lin Li,Jefersson A. dos Santos,Ricardo da Silva Torres +9 more
TL;DR: This paper describes the approaches used by the MultiBrasil team for the Multimedia Satellite Task at MediaEval 2017, which employs neural networks for end-to-end learning for both disaster image retrieval and flood-detection in satellite images.