“CAT: A Graphical User Interface for Visualization and Level-of-Detail Control for Large Scale Image Collections”

This thesis proposes CAT (Clustered Album Thumbnail), a technique for clustering and browsing large number of images. It also provides a user interface for controlling the level of details. As a preprocessing, CAT first hierarchically clusters images and selects representative images for each cluster. And then, CAT visualizes the tree structure applying a hierarchical data visualization technique HeiankyoView. As a characteristic of CAT, it provides a graphical user interface for the zooming operation to effectively browse images. It selectively displays representative images while zooming out, or individual images while zooming in, by the mouse operation. This feature realizes high frame rates, and display of adequate number of images.