Art(-history) and Artificial Intelligence - Which Questions are Most Pressing?

Digital Humanities Questions & Answers » Applications, Tools, Formats

Art(-history) and Artificial Intelligence - Which Questions are Most Pressing?

(4 posts) (3 voices)

Asked 3 years ago by chris.abyi@gmail.com
Latest answer from Ethan Gruber

Tags:

No tags yet.

chris.abyi@gmail.com
Member
Dear community,

I'm just about to start my dissertation which will last 6 months. I've got a background in History of Art and Computer Science and already accomplished a few projects in the area of digital humanities, especially dealing with graph-theoretical and network-analysis problems.

I want to focus my dissertation on a question located in the history of art / art / science of art that involves Artificial Intelligence. I thought it might be worth a try to ask in this forum which problem you consider most pressing in this area, or if there is any comprehensive documentation of open problems.

Thanks for your feedback and
all the best!

Christian
Tweet this question
Posted 3 years ago Permalink
Stéfan Sinclair
Admin

Hi Christian, Did you happen to see this post already (not quite the same as your message, but with some useful pointers I think)? http://digitalhumanities.org/answers/topic/computational-analysis-of-perspective-in-paintingsart

Posted 3 years ago Permalink
chris.abyi@gmail.com
Member

@Stéfan, thanks for the reference. Issues of perspective and image analysis are indeed interesting. Nevertheless, I'm looking for further issues involving artificial intelligence. What about digital technologies in the museum environment? I read Tallon/Walker, but technology progresses very quick and I feel out of date. Any ideas about the role of A.I. in this setting or references to overviews of work being done in the field are highly appreciated!

Posted 3 years ago Permalink
Ethan Gruber
Member

There are usually a number of AI-related papers presented at humanities/technology conferences like Computer Applications in Archaeology and VAST. There was recently a paper on automatic classification of Australian rock art at CAA, but I've seen other papers on classification of Roman coins (based on both 2D images and 3D scans) and laser scanned 3D objects, like the Forma Urbis project. I think there are a lot of avenues for AI in art history. Classification of objects is an often time-consuming task that could be made much faster by automated processes, for example, categorization of Greek pottery based on styles of drawing. I'm frankly not very sure you can accomplish writing sophisticated algorithms for this sort of problem in six months, but you could have a go at it. For example, in order to write AI for attributing Greek pottery based on the drawing styles of painters, you have to have a fairly robust understanding of how the task is accomplished manually, which is to say, you have to understand the styles yourself.

Posted 3 years ago Permalink

RSS feed for this topic

Reply

You must log in to post.