Context-Infused Visual Grounding for Art

TL;DR

This paper introduces CIGAr, a visual grounding method tailored for artworks that leverages textual descriptions during training, along with a new dataset, Ukiyo-eVG, to improve object localization in art images.

Contribution

The paper proposes CIGAr, a novel visual grounding approach that incorporates artwork descriptions as context, and introduces Ukiyo-eVG, a new annotated dataset for phrase-grounding in art.

Findings

CIGAr outperforms existing methods on art datasets.

Ukiyo-eVG dataset provides high-quality annotations for art grounding.

Achieved state-of-the-art object detection results in artwork datasets.

Abstract

Many artwork collections contain textual attributes that provide rich and contextualised descriptions of artworks. Visual grounding offers the potential for localising subjects within these descriptions on images, however, existing approaches are trained on natural images and generalise poorly to art. In this paper, we present CIGAr (Context-Infused GroundingDINO for Art), a visual grounding approach which utilises the artwork descriptions during training as context, thereby enabling visual grounding on art. In addition, we present a new dataset, Ukiyo-eVG, with manually annotated phrase-grounding annotations, and we set a new state-of-the-art for object detection on two artwork datasets.