Event Summary

Presentation

Access to non-text content has long been achieved by analysis of textual metadata attached to the media records. This is obviously a primitive mode of access since it is limited by the ability of the metadata to capture the essence of what is being seen. Since 1995, researchers have been developing technology that allows access to visual content at a deeper level through examination of visual features in the digital space like colour, shape and motion. This body of work is known as "Content based analysis" and it has rapidly become a mature field.

 

Speakers

Anil Kokaram, Fellow of Trinity College, Dublin