What is it about?
In the history of art, drawings often present more complex histories than better-documented “finished” works, like paintings. Drawings served not only as vehicles of artistic expression, but also as preparatory studies, experiments with various techniques and media, and to hone drafting skills. Often unsigned and treated as utilitarian, drawings rarely enjoyed the care and respect accorded to painted works. Frequently executed using multiple media and scarred by mishandling over the years, drawings present a particular challenge to the authenticator – especially if the “authenticator” is a computer. While many machine learning efforts have been devoted to the analysis of paintings, little attention has been paid to drawings.
Featured Image
Photo by Andrey Novik on Unsplash
Why is it important?
We have developed a deep learning technique tailored to drawings. Our technique is robust to the age and wear of a drawing as well as the possibility that it contains marks made with multiple drawing media. We obtained classification accuracies exceeding 90% on a curated set of drawing images. In particular, we focused on works attributed to Raphael, as well as drawings by his admirers, imitators, and forgers.
Perspectives
The increasing difficulty of obtaining expert opinion on questions of authenticity is well-documented; art experts fear loss of reputation as well as loss in the courtroom. Our system offers the possibility of a convenient ‘first look’ at attribution: one that may suggest the utility of professional expertise and perhaps even encourage otherwise reticent experts to venture their opinions.
Steven Frank
Art Eye-D Associates; Med*A-Eye Technologies
Read the Original
This page is a summary of: Analysis, Attribution, and Authentication of Drawings with Convolutional Neural Networks, International Journal of Arts and Technology, January 2022, Inderscience Publishers,
DOI: 10.1504/ijart.2022.10050958.
You can read the full text:
Contributors
The following have contributed to this page
