Abstract
Due to the recent spread of low-cost and high-performance digital cameras and smart phones, an individual can easily possess a large collection of image data, and thus we often see image data flood on the Web. However, when performing image browsing and search an unstructured set of image data, point of view does not always remain consistent with respect to the subject, which may possibly disrupt a cognitive map of the user. In order to tackle the problem, we proposed a method to organize a collection of input images into a tree-of-graph structure in terms of the coverage (scope) of the images. Furthermore, we developed a novel image viewer tailored for intercommunicating between the different layers. In this article, we also report an application of the viewer to browsing of travel photographs, where the users are allowed to perform spatiotemporal browsing of travel photographs selectively.
