Time Quilt: Scaling up Zoomable Photo Browsers for Large, Unstructured Photo Collections

Conference on Human Factors in Computing Systems |

Published by ACM New York, NY, USA

In the absence of manual organization of large digital photo collections, the photos’ visual content and creation dates can help support time-based visual search tasks. Current zoomable photo browsers are designed to support visual searches by maximizing screenspace usage. However, their space-filling layouts fail to convey temporal order effectively. We propose a novel layout called time quilt that trades off screenspace usage for better presentation of temporal order. In an experimental comparison of space-filling, linear timeline, and time quilt layouts, participants carried out the task of finding photos in their personal photo collections averaging 4,000 items. They performed 45% faster on time quilt. Furthermore, while current zoomable photo browsers are designed for visual searches, this support does not scale to thousands of photos: individual thumbnails become less informative as they grow smaller. We found a subjective preference for the use of representative photos to provide an overview for visual searches in place of the diminishing thumbnails.