The Close-up Cloud
Visualizing Details of Image Collections in Dynamic Overviews
This paper introduces a visualization technique designed to uncover iconographic patterns prevalent within a collection while at the same time allowing close viewing of these particular details. Challenging an institutionalized understanding of overview and detail as inherently opposed, the intention of this research is to develop a visualization method that accounts for the iconographic abundance of a collection and encourages its casual exploration. Expanding digitization efforts have led to a growing number of rich cultural heritage datasets that are successively being published online. At the same time scholars are exploring the potential of computational methods to expand the scale and scope of art history. In this context, data visualization is often equated with a distanced perspective diminishing the intricate and intriguing details of individual artifacts. In collaboration with a museum of applied and decorative arts, we have devised a novel interface concept for the exploration of image collections such as historical glass plate negatives. Inspired by photographic plates on a light table, the resulting Close-up Cloud translates the art historical method of close viewing into the digital by combining it with a dynamic representation of quantitative iconographic patterns across an entire image collection.