Objectives:
The aim of this course, together withitsparallelsubjectDeep Learning for Computer Vision1,is to help the studentunderstandandapplythetheoretical and practicalfundamentals behind deep learningtechniques applied toimage analysis and synthesis.
Deep learningtechnologyrepresentsthecurrentstate-of-the-art forimage processing and computer vision applications,andisone of the hottest researchand industrialtopic nowadays,which is in constant development.This course intends to provide the student with the necessary tools tobe ready to apply currentcomputer visiontechnology in the research and industrial fields, andinteriorize the basic concepts thatwill allow forthe monitorization of the future development of thetechnology.
Specifically, this course covers: i)thefundamentals and development ofcoreimage deep learning architectures,which areprevalent inalmost any computer vision pipeline,ii)andbasic mechanisms for generative architectures,iii) the understanding and application ofvisual models learning from natural language supervisionandthe associatedimage,textandcoordinatedrepresentation learning techniques,and iv)an introduction tothe social and environmental challenges associateddeep learningcomputer vision;includinga discussion onthe implications of regulations such as the EU AI Act, addressing sustainability,interpretability,andreliability challenges of these models.
More information about this course is available HERE.