Project Description

Our visualization's goal is to create the “ideal” layout of a photo collection, arranging the photo thubmails in such a way as to match people's underlying cognitive process of identifying groups of related photos. (Or as close to ideal as we can get). The data behind this layout will be a dynamically calculated value of “closeness” between any two photos, based on similarities in the photos' metadata. The relative attractive weighting of the various types of metadata will be adjustable through the user interface, allowing a live exploration of the photo collection's visualization.

Our hope is that the interface will be intuitive enough as to allow people to adjust the weights to maximize the similarity between the on-screen grouping of photographs and how an individual would classify or tag their photo collection. This technique may be applicable to the larger problem of photo tagging and annotation. The act of using the grouping to easily tag photos, and dynamically recalculating the weighting based on the grouping, is out of scope of this project but may be a future direction for the team.

Team Members
Carrie Burgener ~ carrie@sims
Benjamin Hill ~ bhill@sims
David Hong ~ dkhong@sims