This page summarizes all the activities we have developed in the context of CAPES STIC-AmSud project (2019-2020).
Optimized Deep Learning based Representations for Computer Vision Problems (DLRCV)
Abstract
The main goal of this project is to investigate deep learning models to learn optimized descriptions for computer vision problems such as handwritten recognition, document spotting, image retrieval, 3D shape analysis, among others . To this end, we will need to achieve the following specific objectives: to investigate methods for reducing dimensionality of deep features in problems related with image retrieval; to investigate deep learning model to learn binary descriptions from images like sketches;to investigate end-to-end trained deep learning models for cheque amount recognition. to investigate deep learning representations for signature verifications; to investigate deep learning models to learn the more representative views of 3D objects in the context of 3D shape analysis; to investigate optimized deep learning representations for document spotting.
Institutions
- Université de Rouen Normandie (URN, Rouen, France)
- Pontificia Universidade Católica do Paraná (PUCPR, Curitiba, Brazil)
- Universidade Federal do Paraná (UFPR, Curitiba, Brazil)
- Universidade Federal de Pernambuco (UFPE, Recife, Brazil)
- Pontificia Universidad Católica del Perú (PUCP, Lima, Peru)
- Universidad de Santiago de Chile (USACH, Santiago, Chile)
- ORAND S.A. (Santiago, Chile)
Team
- Prof. Alceu de Souza Britto Jr (PUCPR/Brazil, international coodinator)
- Prof. Luiz Eduardo S. Oliveira (UFPR/Brazil)
- Prof. George Cavalcanti (UFPE/Brazil)
- Prof. Laurent Heutte (Université de Rouen/France)
- Prof. Caroline Petitjean (Université de Rouen/France)
- Prof. Violeta Chang (Univerdidad de Santiago de Chile)
- Prof. Iván Sipirán Mendoza, PhD (PUCP/Peru)
- Profa. Analí Alfaro Alfaro, PhD (PUCP/Peru)
- José Saavedra, Dr. (ORAND S.A./Chile)
- Juan Barrios, Dr. (ORAND S.A/Chile)
- Camila Alvarez, MSc (ORAND S.A./Chile)
- Andrés Ferrada (Master student, University of Chile/Chile)
- Teruo Matos Maruyama (PhD student, UFPR/Brazil)
- Marcelo Trierveiler Pereira (PhD student, PUCPR/Brazil)
- Bruna Delazari (PhD student, PUCPR/Brazil)
- Paulo Roger Gomes Cordeiro (UFPE/Brazil)