PhD scholarship in Visual motifs identification and comparative image learning

Film and photography, art, and creative communication frequently include visualmotifs. That is, identifiable compositions and patterns that image-makers use to express things visually. Experts have identified a considerable set of such visual motifs appearing in the aforementioned creative areas. The goal of this project is to automatically learn and identify those visual motifs, enhancing the capacity of image analysis with the complex comparative models of iconography. A direct application of this work could be the development of legacy based tools for image analysis and improvement that go beyond formal patterns (like shot scale or lightning) and provide aesthetic and artistic enrichment by identifying and suggesting visual motifs across media. For example, being able to identify representations of such a universal motif for the representation of suffering that is La Pietà, not only in painting or sculpture, but also in photojournalism and cinema.
This PhD will be part of the VIMICIL project, which is an interdisciplinary project where researchers in computer vision and machine learning collaborate with researchers in audiovisual communication, specialised in cinema and the visual arts. The goal of the PhD is
to design a learning-based model capable of identifying a set of visual motifs in images and videos. We plan to work with a dataset of visual works that has been developed by experts in narrative, film studies, comparative media and art historians.

This position is co-funded by the PhD fellowship program of the Department of Information and Communication Technologies at Universitat Pompeu Fabra (DTIC-UPF), and the María de Maeztu Strategic Research Programme at DTIC-UPF on Artificial and Natural Intelligence for ICT and beyond. Its benefits and conditions are available at: https://www.upf.edu/web/etic/predoctoral-research-contracts.

Open position for a predoctoral researcher:

This position includes a teaching commitment load of 45 hours per academic year.

Topic: Automatic visual motifs identification and comparative image learning

Requirements:

1. Master degree in computer vision, computer science, data science, artificial intelligence orsimilar.
2. Experience in computer vision and deep learning.
3. Interest in and knowledge ofcinema and art is also desirable.

Research field, educational level, other requirements:
Admission in the PhD program of the Department of Information and CommunicationTechnologies at UPF is a prerequisite to enjoy the contract.

Starting date (planned): October 2023

Application deadline: 30th June 2023

Gross monthly salary: 1.680€. (To increase to 2020€ during the fourth year of the PhD).

For application or further information, contact:

• Coloma Ballester, IMVA-group, coloma.ballester@upf.edu, Gloria Haro, IMVA-group,
• gloria.haro@upf.edu, Manuel Garin, CINEMA-group, manuel.garin@upf.edu

More information about the María de Maeztu Strategic Research Programme at DTIC-UPF on Artificial and Natural Intelligence for ICT and beyond: https://www.upf.edu/web/mdm-dtic .