Abstract

Pictures taken by journalists for distribution and for inclusion in stock photo collections are often enriched with metadata. One key aspect of such photos is that they focus largely on events and feature celebrities and other public figures. Such photographs may provide interesting insights into how such public gures are related to each other, in terms of the events they attend and in their social proximity in terms of how often they are photographed together. In this paper, we study a corpus of approximately 9 million stock photographs taken over a 10 year period and, using metadata from these photos, we extract a social network from co-appearance of public gures in events depicted in the photographs. We exploit this latent social information and combine it with the rich semantic metadata attached to the images to explore the possibility of predicting attendees at future events. Our results show promising performance for this event attendance prediction task.