A Neural Network Looks at Leonardo's(?) Salvator Mundi

TL;DR

This paper employs convolutional neural networks to analyze the attribution of Leonardo da Vinci's Salvator Mundi, aiming to identify forgeries and clarify authorship controversies through visual analysis.

Contribution

It introduces a CNN-based method trained on Leonardo's works and comparable artworks to assist in attribution and forgery detection.

Findings

CNN can differentiate Leonardo's style from others

The system aids in identifying potential forgeries

Limitations exist due to the few extant Leonardo paintings

Abstract

We use convolutional neural networks (CNNs) to analyze authorship questions surrounding the works of Leonardo da Vinci -- in particular, Salvator Mundi, the world's most expensive painting and among the most controversial. Trained on the works of an artist under study and visually comparable works of other artists, our system can identify likely forgeries and shed light on attribution controversies. Leonardo's few extant paintings test the limits of our system and require corroborative techniques of testing and analysis.