Facial recognition technology has established that the mystery e de Brécy Tondo painting was almost certainly painted by Raphael.

New facial recognition technology has been used to examine a painting known as the de Brécy Tondo, leading researchers to conclude that it is highly likely to be a masterpiece by Raphael. The technology was developed by Professor Hassan Ugail, of the University of Bradford, and uses a deep neural network to identify patterns in images with greater accuracy than the human eye. (University of Nottingham)

Using this AI, researchers from the University of Nottingham and University of Bradford compared faces in the Tondo painting to those in Raphael’s Sistine Madonna altarpiece and found that they were identical, indicating that the two paintings were created by the same artist.

The similarity between the Madonnas in the two paintings was found to be 97%, while the similarity between the Child in both works was 86%. A similarity more than 75% is considered to be identical.

Further analysis using Scanning Electron Microscopy and Raman spectroscopy confirmed that the painting’s characteristics were typical of Renaissance practice, adding weight to previous research that found pigments in the Tondo to be consistent with early Renaissance work.

The de Brécy Tondo was bought in 1981 by George Lester Winward, who established the de Brécy Trust Collection in 1995 to preserve his collection of art spanning from the sixteenth to the nineteenth century.

“The Trust is absolutely delighted that this new scientific evidence confirms the Raphael attribution of the Tondo, following Raman spectroscopy analysis of its pigments by Professor Howell Edwards confirming its dating to the Renaissance period,” said Timothy Benoy, Honorary Secretary, the de Brécy Trust. “It illustrates very forcibly the increasing value of scientific evidence in the attribution of a painting.”

The research paper, titled Deep Facial Features for Analysing Artistic Depictions, was presented at the International Conference SKIMA2022 in Phnom Penh, and is due to be published later this month.

The use of facial recognition technology in art analysis could have far-reaching implications for the attribution of paintings in the future.

Related Story:  AI Simulates Long Lost Portions of Rembrandt’s Masterpiece

Image Credit: University of Nottingham