Computer vision as critical practice:how DH can teach computers to say what they see

Computer vision has made fundamental advances in recent years, but has only just begun to be adopted by digital humanists. This paper will outline some of its humanities applications, including how computational analysis relates to established descriptive practices. Examples will include popular print, early photography and scientific illustration within fields such as book history, conservation research and visual studies.

The paper will both demonstrate what can be done now with open-source tools, and reflect on what remains to be done to truly establish a home for computer vision within the digital humanities.

All welcome and there will be drinks and discussion after the talk. Attendance is free but we kindly ask that you register for the event.

Speaker

Giles Bergel is Digital Humanities Research Ambassador in the Visual Geometry Group, in the Department of Engineering Science, University of Oxford. 

With a background in book history, he works on the computational analysis of both images and text, particularly early printed books and broadside ballads. His current project is Seebibyte: Visual Search for the Era of Big Data.