Annotation of cultural heritage documents based on XML dictionaries and data clustering

Abstract

Cultural heritage forms the local and national identities. It shapes relationships between neighbors and other communities around the world. The sweet wine named "Commandaria" is part of Cypriot heritage and currently holds a protected destination of origin within European Union, USA and Canada. In the framework of the Commandaria project we managed to gather an enormous amount of data, related to Commandaria wine, corresponding to photographs, scanned documents and videos. The need of a method for efficient retrieval of these data based on their actual content was mandatory. The data were appropriately indexed through a multilevel labeling scheme allowing access from various modalities and for a variety of applications. Despite the huge efforts for automatic characterization and classification human intervention is the only way for reliable multimedia data annotation. Manual data annotation is an extremely laborious process and efficient tools developed for this purpose can make, in many cases, the true difference. In this paper we present the CulHIAT, a cultural heritage item annotation tool, which uses structured knowledge, in the form of XML dictionaries, combined with a hierarchical classification scheme, to attach semantic labels to image and video segments at various levels of granularity. Finally, XML dictionary creation and editing tools are available during annotation allowing the user to always use the semantic label she/he wishes instead of the automatically created ones