While institutions like the Cantonal University Library of the Canton of Fribourg (BCU) traditionally offer services such as loans and consultations, their role extends to preserving and making accessible the cantonal heritage. However, transitioning from preservation to public engagement requires a specific mediation effort to ensure sufficient appeal. This challenge forms the central mission of the EPFL+ECAL Lab.
Our endeavor to enhance the value of the BCU’s patrimony began with the Èvokâ project, where we encountered challenges in interlinking archives and generating appeal based on the existing metadata. To address this, we aim to strengthen the connections between visual archives and historical contexts. Our proposal involves utilizing the archives of Fribourg’s newspapers to enhance the metadata associated with photographic and postcard collections, thereby creating richer, more meaningful links.
The primary focus and challenge of the proposed project is to investigate algorithms that can extract information from both visual and textual content and link them together. This project aims to use Vision-Language Models to make connections. It will do this by using the models’ ability to understand and connect pictures with related contexts.
The involved student will be part of the EPFL+ECAL Lab team conducting the design research project composed of engineers, designers, and psychologists. Ultimately, the results of the project will be integrated within the Èvokâ prototype.
Key Domains
- VLM
Prerequisites
- Native French speaker
- Some knowledge of machine learning, deep-learning, fine-tuning LLM.
- Proficiency in Python.
Contact