EUFCC-CIR: a Composed Image Retrieval Dataset for GLAM Collections

TL;DR

This paper introduces EUFCC-CIR, a large dataset for Composed Image Retrieval in cultural heritage collections, enabling better AI-driven exploration of GLAM archives.

Contribution

It provides a novel, extensive CIR dataset tailored for Digital Humanities, filling a gap in resources for cultural heritage image retrieval.

Findings

EUFCC-CIR contains over 180K annotated triplets.

The dataset demonstrates unique qualities compared to existing CIR datasets.

Zero-shot CIR baselines show promising performance on EUFCC-CIR.

Abstract

The intersection of Artificial Intelligence and Digital Humanities enables researchers to explore cultural heritage collections with greater depth and scale. In this paper, we present EUFCC-CIR, a dataset designed for Composed Image Retrieval (CIR) within Galleries, Libraries, Archives, and Museums (GLAM) collections. Our dataset is built on top of the EUFCC-340K image labeling dataset and contains over 180K annotated CIR triplets. Each triplet is composed of a multi-modal query (an input image plus a short text describing the desired attribute manipulations) and a set of relevant target images. The EUFCC-CIR dataset fills an existing gap in CIR-specific resources for Digital Humanities. We demonstrate the value of the EUFCC-CIR dataset by highlighting its unique qualities in comparison to other existing CIR datasets and evaluating the performance of several zero-shot CIR baselines.