UGESCO - A Hybrid Platform for Geo-Temporal Enrichment of Digital Photo Collections Based on Computational and Crowdsourced Metadata Generation

摘要

The majority of digital photo collections at museums, archives and libraries are facing (meta) data problems that impact their interpretation, exploration and exploitation. In most cases, links between collection items are only supported at the highest level, which limits the item's searchability and makes it difficult to generate scientific added value out of it or to use the collections in new end-user focused applications. The geo-temporal metadata enrichment tools that are proposed in this paper tackle these issues by extending and linking the existing collection items and by facilitating their spatio-temporal mapping for interactive querying. To further optimize the quality of the temporal and spatial annotations that are retrieved by our automatic enrichment tools, we also propose some crowdsourced microtasks to validate and improve the generated metadata. This crowdsourced input on its turn can be used to further optimize (and retrain) the automatic enrichments. Finally, in order to facilitate the querying of the data, new geo-temporal mapping services are investigated. These services facilitate cross-collection studies in time and space and ease the scientific interpretation of the collection items in a broader sense.