Collaborative event annotation in tagged photo collections

Abstract

Events constitute a significant means of multimedia content organization and sharing. Despite the recent interest in detecting events and annotating media content in an event-centric way, there is currently insufficient support for managing events in large-scale content collections and limited understanding of the event annotation process. To this end, this paper presents CrEve, a collaborative event annotation framework which uses content found in social media sites with the prime objective to facilitate the annotation of large media corpora with event information. The proposed annotation framework could significantly benefit social media research due to the proliferation of event-related user-contributed content. We demonstrate that, compared to a standard "browse-and-annotate" interface, CrEve leads to a 19% increase in the coverage of the generated ground truth in a large-scale annotation experiment. Furthermore, the paper discusses the results of a user study that quantifies the performance of CrEve and the contribution of different event dimensions in the event annotation process. The study confirms the prevalence of spatio-temporal queries as the prime option of discovering event-related content in a large collection. In addition, textual queries and social cues (content contributor) were also found to be significant as event search dimensions. Finally, it demonstrates the potential of employing automatic photo clustering methods with the goal of facilitating event annotation.