IMAGE ANALYSIS OF FINE ARTS AND CULTURAL HERITAGE

This short course is a part of the Ph.D. program in Computer Science. It aims at the exploration of the diverse tasks and deep-learning techniques utilized in AI research for visual art understanding.

Students will be able to demonstrate a comprehensive understanding of the tasks used in AI research for understanding visual art. They will gain familiarity with state-of-the-art approaches, key concepts, and tools employed in the automatic analysis of fine art and cultural heritage.

Foundations of machine learning. Basic knowledge deep learning. Use of computational thinking. Python Coding.

The program is provisional and is subject to change. Its contents may include a selection from the following potential topics:
- Main research trends, art collections as data sources, distant viewing, and close reading
- Automated classification of artworks based on artist, style, genre
- Utilization of deep learning methodologies for detecting objects (people, faces, motifs, objects, materials, characters) within artworks
- Retrieving visually linked images in painting collections using deep learning approaches.
- Visual question-answering methods and generating captions for artworks.
- Creation of art through Generative Adversarial Networks (GANs).

Scientific papers

A presentation on a scientific article or a software development project related to the topics of the course

Theoretical lessons describe the various concepts and methods; active learning and open discussions.

English

oral