This project is developing and testing computer based methods for extracting image object metadata from partners’ collections, storing it in a data warehouse where it can be searched for potential matches. Results are stored in a second “links” database which is interrogatable by a “widget” embedded in partners’ Web sites. When visitors to any one of these sites visit an individual image record, the widget will offer them a selection of hyperlinks to potentially related records held in other partners’ collections.
The FuzzyPhoto project is a two year AHRC funded research project (AH/J004367/1) hosted by the De Montfort University Photographic History Research Centre in collaboration with the Centre for Computational Intelligence . Project partners are Birmingham Central Library, the British Library, the Musée D’Dorsay and the Louvre, Paris, the Metropolitan Museum, New York, the National Media Museum, St Andrews University and the V&A. The project started in November 2012, ending in November 2014.
Fuzzyphoto is centred around Photographs Exhibited in Britain 1839-1865 (peib.dmu.ac.uk) and Exhibitions of the Royal Photographic Society 1870-1915 (erps.dmu.ac.uk), a pair of databases that together comprise the single most comprehensive record of British photographic exhibitions in the nineteenth and early twentieth centuries. These early exhibition catalogues were often devoid of pictures, relying instead on written descriptions of the image or artists’ impressions.
As image collections are increasingly being published online there is an opportunity to try to match images of photographs with these historical exhibition catalogue entries. Making precise associations between exhibit records and images is difficult because photographs were commonly exhibited/published more than once, at different times, with different titles and even by different people. Also collection records in museums galleries and archives are often incomplete, imprecise and inaccurate.
The FuzzyPhoto project is developing computer based methods for extracting image object metadata from partners’ collection records, storing it in a data warehouse where it can be searched offline for potential matches. The results are stored in a second, “links”, database which is interrogatable by a “widget” embedded in partners’ Web sites. When visitors to any one of these sites drill down to the level of an individual image object, the widget displays a selection of hyperlinks to potentially related records held in other partners’ collections. For descriptions of how this is done, see methods, reports and publications. For a description of how the project was organised, see management.