1 Introduction The rapidly increasing use of digital cameras is caus- ing a corresponding increase in the number and size of personal digital photo collections. These collec- tions routinely contain thousands of photos and require effective interfaces that facilitate browsing, manipulation, and sharing. Studies of user needs have found that users exploit various aspects of digi- tal photography in usage scenarios such as sharing photos with friends and retrieving photos of impor- tant events such as birthdays and weddings (Kuchin- sky et al. 1999; Graham et al. 2002; Schiano et al. 2002; Shen et al. 2002). Before sharing, people typi- cally classify photos into those they wish to share versus those they intend to leave “in the shoebox.” Thus, there is a demand for powerful tools to help users organize, classify, and browse their collections. There are already many commercial and research applications supporting the organization of digital photos (ACD Systems; Apple Computer; Bederson 2001; Canon; Kang and Shneiderman 2000; Kuchin- sky et al. 1999). While our application shares fea- tures with some of them, our goal has been to make organizing and browsing photos simple and quick, while retaining scalability to large collections. To that end, we have concentrated on areas that improve the overall experience without neglecting the mun- dane components of a digital photo organization application. We facilitate organizing and viewing large photo collections by automatically dividing photos into meaningful episodes or events, such as a birthday party or a trip. Though the events are auto- matically detected initially, the resulting boundaries can be manually adjusted if desired. Our application presents photos in a vertically scrollable light table which shows thumbnails of the user’s entire photo collection, with markers indicat- ing the start of each event. A tree view for displaying events and other attributes (people, places, etc.) can be used to scroll the light table to a selected event, to sort photos by different categories, or to filter the set of visible photos to show only a particular category. We fine-tuned our application by testing it with photo collections from six different users each with 500 to 1500 photos. In two of the collections, the photographers organized the photos in directories representing events. We used that information as ground truth for evaluating our event detection algo- rithm. One collection contained photos taken by sev- eral photographers at the same event made us aware of issues with mixing scanned photos and photos from cameras set to different time zones. To test scal- ability, we used our application with the approxi- mately 4300 photos from all six collections simultaneously. In the next section, we review several existing applications and algorithms for organizing digital photos and compare them to our approach. We then describe the automatic techniques and user interface mechanisms in our application. Afterwards, we dis- cuss experiences encountered during the develop- ment and use of the application. We conclude with a discussion of future directions. Simplifying the Management of Large Photo Collections Andreas Girgensohn, John Adcock, Matthew Cooper, Jonathan Foote & Lynn Wilcox FX Palo Alto Laboratory, 3400 Hillview Avenue, Palo Alto, CA 94304, USA {andreasg, adcock, cooper, foote, wilcox}@fxpal.com Abstract: With digital still cameras, users can easily collect thousands of photos. Our goal is to make organizing and browsing photos simple and quick, while retaining scalability to large collections. To that end, we created a photo management application concentrating on areas that improve the overall experience without neglecting the mundane components of such an application. Our application automatically divides photos into meaningful events such as birthdays or trips. Several user interaction mechanisms enhance the user experience when organizing pho- tos. Our application combines a light table for showing thumbnails of the entire photo collection with a tree view that supports navigating, sorting, and filtering photos by categories such as dates, events, people, and locations. A calendar view visualizes photos over time and allows for the quick assignment of dates to scanned photos. We fine-tuned our application by using it with large personal photo collections provided by several users. Keywords: Digital photo collections, automatic event detection, user-centered design.