Authentication
Machine learning and image
analysis tools can be used to assist art experts in the
authentication of unknown or disputed paintings. In reexamining
recently successful methods, we find that varying image quality
may have acted as a confounding factor, artificially improved
the results. To carry out an independent test, we gathered a new
“ground truth” data set in which originals and copies are known
and image acquisition conditions are uniform. We use Hidden
Markov Trees (HMTs) to model the wavelet coefficients of these
paintings; the resulting HMT model parameters are used as input
features for state-of-the-art machine learning algorithms to
distinguish copies
and originals.
A Stylistic Analysis: Dating
and Finding Distinguishing Features
Wavelet transforms and machine
learning tools can be used to assist art experts in the
stylistic analysis of paintings. A dual-tree complex wavelet
transform, Hidden Markov Tree
modeling and Random Forest classifiers are used here for a
stylistic analysis of Vincent van Gogh’s paintings with results
on two stylometry challenges that concern “dating, resp.
extracting distinguishing features”.
|