Current Projects
Dense Labeling of Remote Sensing Images in the Wild
This project address the challenges for the effective use of supervised learning in dense pixel ...
Pattern recognition using small annotated data sets
This project proposes the development of new approaches to deal with pattern recognition in ...
Deep Representations for Large-Scale Geo-Mapping
This project aims to address the problem of pattern recognition for the creation of thematic maps ...
Past Projects
AI Methodologies For Environmental Monitoring
This research project aims to develop methodologies in the selection process of audit objects ...
Diagnosis Aid Using Chest X-Ray and Deep Learning
The CADCOVID-19 project aims to develop an online system to assist in the diagnosis of COVID-19 ...
Automatic Thematic Maps Using Multiple Sensors
Geographic Information Systems (GIS) are automated systems whose purpose is to store, analyze and ...
Artificial Intelligence Systems for External Control
As políticas de proteção ambiental e climática requerem complexas estruturas de governança ...
Partners
Our team has conducted/participated in various consulting and R&D activities with companies and important public institutions. We have also wide experience in Continuing Education Programs (extension school) for industry professionals.
News
Conheça um pouco mais sobre o nosso laboratório:
https://www.instagram.com/reel/CdGA6EuDcGY/?utm_source=ig_web_button_share_sheet
Semantically segmenting images using SOT techniques
often require large amounts of labeled data, which is laborious to produce.
Our recent work leverages meta-learning to segment using few-shot examples of sparsely labeled data, which are easier to acquire
https://ieeexplore.ieee.org/document/9744585