Abstract
We demonstrate how movements of multiple people or objects within a building can be displayed on a network representation of the building, where nodes are rooms and edges are doors. Our representation shows the direction of movements between rooms and the order in which rooms are visited, while avoiding occlusion or overplotting when there are repeated visits or multiple moving people or objects. We further propose the use of a hybrid visualization that mixes geospatial and topological (network-based) representations, enabling focus-in-context and multi-focal visualizations. An experimental comparison found that the topological representation was significantly faster than the purely geospatial representation for three out of four tasks.
