Public Datasets

  1. LeMe-PT: A public corpus of Leaflets of Medicine – a corpus of leaflets collected from Infarmed and annotatted for natural language processing and data extraction.
  2. MoLa RGB CovSurv: This repository includes the complete dataset used for the training, validation and testing tasks, in order to detect the presence or absence of mask by people in public areas.
  3. MoLa IR CovSurv: This repository includes the complete dataset used for the training, validation and testing tasks, in order to detect the presence ou absence of mask and glasses on people, and detect the caruncle zone of people’s eyes (both on a thermografic context).
  4. Target image dataset for camera calibration: This is a dataset with images of a calibration target, taken with a variety of different cameras, from DSLR, compact, GoPro, phone cameras and Raspberry Pi. For the Cannon 6D, different lenses were used.
  5. Fire occurrence dataset with geographic data of burned areas, geographic data of land occupation, and meteorological conditions: This repository contains a dataset of forest fires in Portugal in 2017 and associated metadata of land use/land cover, digital elevation model and weather.
  6. Functional prototype for forecasting burned areas through the processing of geographic data on land use and meteorological conditions.
  7. Dataset MoLa R10k InCar Dataset Pt1: A dataset of real infrared images of car interiors and passenger joint position for computer vision (part 1).
  8. Dataset MoLa R10k InCar Dataset Pt2: A dataset of synthetic infrared images of car interiors and passenger joint position for computer vision (part 2).
  9. Dataset of ornamental stone images with ground-truth classification for computer vision, in collaboration with the company Mercado da Pedra: A dataset with images and metadata of natural stones.
  10. Video dataset of violent and non-violent scenes with groundtruth classification for computer vision.
  11. Formula 1 car image and video dataset with groundtruth for image classification and object detection with computer vision: A dataset of images of F1 cars for image classification and object detection.
  12. Functional prototype of image classification on FPGA devices (Intel Altera DE1-SoC)
  13. High, medium and low quality football video dataset with groundtruth of player detection and tracking for computer vision: A dataset with video in 3 quality levels with player tracking groundtruth data.
  14. SatSuperRes dataset: dataset with high quality/ low quality image pairs of satellite imagery. Public available under request through the e-mail ailaboratory@ipca.pt.
  15. SatFire3 dataset: dataset with burned areas and images before and after each fire. Public available under request through the e-mail ailaboratory@ipca.pt.
  16. SatForest2 dataset: dataset with Sentinel2 satellite imagery and land use/land cover classification (COS2018). Public available under request through the e-mail ailaboratory@ipca.pt.