Deep Learning and Visual Artificial Intelligence

It focuses on the development of methods that enable image analysis using deep neural networks, including convolutional neural networks (CNNs), recurrent neural networks (RNNs), and visual transformers (ViT).

Among its applications are object recognition, semantic segmentation, which involves dividing an image into semantically significant regions, image classification, anomaly detection, and image generation (GANs). Although very powerful, these neural networks face the problem of interpretability, so we are also working on methods that allow these “black boxes” to explain their behaviour.

In this context, we have developed projects related to vehicle classification, object detection, visual detection of graphical anomalies, satellite image segmentation, programming of a video-guided autonomous robot, and medical image segmentation, among others. We also offer courses focused on Deep Learning and its practical applications.