The main problems presented in most image processing applications are a large number of features and the difficulty in dealing with the concurrent variations in position, orientation, and scale.
For these situations, some architectures of Deep Neural Networks offer greater flexibility and lower data requirements for training compared to traditional neural networks.
The aim of the project is exploring, implementing and using some Deep Neural Networks architectures, where the original problem can decompose into sub-problems at different levels of abstraction; In Computer Vision, little geometric variations, like edge detection, are extracted from the pixels, then local shapes are obtained until reaching the objects level.
Image Analysis Using Neural Networks
We are always interested in new challenges and looking for new motivated people to work with.
If you are interested in colaborated with us, please contact us.
GITIA was founded on 2008 in order to promote and apply recent research works on advanced technologies to different of domains.
Our main application areas cover collaborative work, decision-helping tools and distributed problem solving.