LIBER Digital Cultural Heritage Forum Workshop: State of the Art in Image Recognition
This workshop will explore why the use of images for research purposes has not yet received as much attention from librarians as full text searching. This may be to do with copyright issues and/or the limitations that exist in indexing and searching visual data, or indeed the difficulties with images in applying tools for classification and recognition.
Potentially, content-based image searching has the huge advantage that the user is not dependent on metadata or cataloguing data to discover a particular image. Traditionally, content-based image comparison checks data on a pixel-bypixel basis and searches a database or the Internet for like images (using technology to mirror and allowing for fuzziness). Subsequently, the need for object recognition in content-based image retrieval has been recognised and developed.
In this workshop we will explore the state of the art in image recognition, especially in terms of searching data, and comparing and classifying images.