(Roughly) Daily

“These are the forgeries of jealousy”*…

Analysis of Leonardo da Vinci’s Salvator Mundi required dividing a high-resolution image of the complete painting into a set of overlapping square tiles. But only those tiles that contained sufficient visual information, such as the ones outlined here, were input to the author’s neural-network classifier.

Is it authentic? Attorney and AI practitioner Steven J. Frank, working with his wife, art historian and curator Andrea Frank (together, Art Eye-D Associates), brings machine learning to bear…

The sound must have been deafening—all those champagne corks popping at Christie’s, the British auction house, on 15 November 2017. A portrait of Jesus, known as Salvator Mundi (Latin for “savior of the world”), had just sold at Christie’s in New York for US $450.3 million, making it by far the most expensive painting ever to change hands.

But even as the gavel fell, a persistent chorus of doubters voiced skepticism. Was it really painted by Leonardo da Vinci, the towering Renaissance master, as a panel of experts had determined six years earlier? A little over 50 years before that, a Louisiana man had purchased the painting in London for a mere £45. And prior to the rediscovery of Salvator Mundi, no Leonardo painting had been uncovered since 1909.

Some of the doubting experts questioned the work’s provenance—the historical record of sales and transfers—and noted that the heavily damaged painting had undergone extensive restoration. Others saw the hand of one of Leonardo’s many protégés rather than the work of the master himself.

Is it possible to establish the authenticity of a work of art amid conflicting expert opinions and incomplete evidence? Scientific measurements can establish a painting’s age and reveal subsurface detail, but they can’t directly identify its creator. That reLeonardo da quires subtle judgments of style and technique, which, it might seem, only art experts could provide. In fact, this task is well suited to computer analysis, particularly by neural networks—computer algorithms that excel at examining patterns. Convolutional neural networks (CNNs), designed to analyze images, have been used to good advantage in a wide range of applications, including recognizing faces and helping to pilot self-driving cars. Why not also use them to authenticate art?

That’s what I asked my wife, Andrea M. Frank, a professional curator of art images, in 2018. Although I have spent most of my career working as an intellectual-property attorney, my addiction to online education had recently culminated in a graduate certificate in artificial intelligence from Columbia University. Andrea was contemplating retirement. So together we took on this new challenge…

With millions at stake, deep learning enters the art world. The fascinating story: “This AI Can Spot an Art Forgery,” @ArtAEye in @IEEESpectrum.

* Shakespeare (Titania, A Midsummer Night’s Dream, Act II, Scene 1)


As we honor authenticity, we might spare a thought for a champion of authenticity in a different sense, Joris Hoefnagel; he died on this date in 1601. A Flemish painter, printmaker, miniaturist, draftsman, and merchant, he is noted for his illustrations of natural history subjects, topographical views, illuminations (he was one of the last manuscript illuminators), and mythological works.

Hoefnagel made a major contribution to the development of topographical drawing. But perhaps more impactfully, his manuscript illuminations and ornamental designs played an important role in the emergence of floral still-life painting as an independent genre in northern Europe at the end of the 16th century. The almost scientific naturalism of his botanical and animal drawings served as a model for a later generation of Netherlandish artists.  Through these nature studies he also contributed to the development of natural history and he was thus a founder of proto-scientific inquiry.

Portrait of Joris Hoefnagel, engraving by Jan Sadeler, 1592 (source)
%d bloggers like this: