Recovery of underdrawings and ghost-paintings via style transfer by deep convolutional neural networks: A digital tool for art scholars

TL;DR

This paper presents a deep learning style transfer method to visualize underdrawings and hidden paintings in art, providing a cost-effective alternative to traditional imaging techniques and offering new insights into artistic works.

Contribution

The authors develop a neural network-based style transfer approach that reveals colors and designs in underdrawings without expensive equipment, aiding art conservation and analysis.

Findings

Successfully applied to Picasso and Leonardo paintings

Reveals colors and designs respecting natural segmentation

Offers a cost-effective alternative to physical imaging methods

Abstract

We describe the application of convolutional neural network style transfer to the problem of improved visualization of underdrawings and ghost-paintings in fine art oil paintings. Such underdrawings and hidden paintings are typically revealed by x-ray or infrared techniques which yield images that are grayscale, and thus devoid of color and full style information. Past methods for inferring color in underdrawings have been based on physical x-ray fluorescence spectral imaging of pigments in ghost-paintings and are thus expensive, time consuming, and require equipment not available in most conservation studios. Our algorithmic methods do not need such expensive physical imaging devices. Our proof-of-concept system, applied to works by Pablo Picasso and Leonardo, reveal colors and designs that respect the natural segmentation in the ghost-painting. We believe the computed images provide insight into the artist and associated oeuvre not available by other means. Our results strongly suggest that future applications based on larger corpora of paintings for training will display color schemes and designs that even more closely resemble works of the artist. For these reasons refinements to our methods should find wide use in art conservation, connoisseurship, and art analysis.