Communications of the ACM
From Brueghel to Warhol: AI Enters the Attribution Fray
Art historians are starting to take advantage of machine learning in their research.
Credit: Eric Monson, Elizabeth Honig
Art historians around the world are starting to utilize machine learning to provide empirical support for theories and ideas previously limited to the subjective eye of the beholders.
For example, University of California, Berkeley researcher Elizabeth Honig used a database of more than 1,500 digitally reproduced Brueghel pictures to train an artificial intelligence (AI) algorithm to pick out identical images of windmills in multiple paintings.
Mathieu Aubry, a researcher at École des Ponts ParisTech in France, uses a technique called unsupervised deep learning, in which the algorithm is shown pictures and finds similarities for itself, to develop more practical applications of AI vision, such as for self-driving cars.
At Rutgers University, researchers are using similar technology to map how artists’ styles are defined and develop over time. The project confirmed that shifts in artistic style could be analyzed and characterized according to binary characteristics.
From Nature
View Full Article
Abstracts Copyright © 2019 SmithBucklin, Washington, DC, USA
No entries found