Visualization of Orientations of Spatial Historical Photographs

Abstract

Historical imagery are an important basis for research in Digital Humanities (DH). Especially art and architectural historians rely on historical photographs that are provided by online media repositories. In general, querying those image repositories is based on metadata. Unfortunately, these are often incomplete, imprecise, or wrong, impeding the search process. Using photogrammetric methods to spatialize the historical imagery, keyword-based search is enhanced by time- and location-dependent browsing methods within a four-dimensional model. The interactive, spatial presentation and exploration of these images opens up new potentials to answer research questions related to art and architectural historical science. One important aspect of the work presented here is to provide visualization methods that present statistical information about image positions, and in particular camera orientations. In addition to heat maps, we present adaptations of methods from flow field visualization to enable the exploration of camera orientations in large numbers of photographic images.