The MIPRCV consortium organizes its first evaluation campaign, based on the selection of some benchmarks available at the website of WP6 few months ago. We invite the members of the consortium to participate in this first campaign. In the following editions, the campaigns will be open to all the scientific community. There will be six main tasks in this first campaign:

• Satellite. Multispectral images in the thermal and visible bands for classification and regression with active learning methods. Organizer: Angeles López (CVDSP-UJI group)
  

• MNIST. The goal is to apply active learning methods to obtain similar results as the reference methods, but using a much less information. Organizer: Ernest Valveny (PR-CVC group)
  

• CD Covers. Database of cover images, designed for studying the problem of unconstrained text detection in complex backgrounds. Organizer: Jordi Vitrià (PR-CVC group)
  

• FKI-PRHLT  BiModalHTR. Database composed by both on-line and off-line word-size segments to apply multimodal fusion techniques. The goal is to classify the bi-modal test samples with the highest accuracy. Organizer: Roberto Paredes (PRHLT-ITI group)
  

• Image relevance feedback. Two image databases are provided to compare different strategies of user interaction in the image relevance feedback context. Organizer: Roberto Paredes (PRHLT-ITI group)
  

• Tarrlab face. Benchmark of pictures of people involving various class labels for  multidimensional classification of faces. Organizer: Luis Baumela (CVL-UGR group)
  

  
  

  Click here to access to the evaluation campaign.