Skip to Content
Computing

A virtual version of da Vinci’s mystery glass orb has helped explain its weirdness

The world’s costliest painting depicts a glass sphere with curious optical properties. Computer scientists figured out what the artist was getting at.
Salvator Mundi by Leonardo Da Vinci
Salvator Mundi by Leonardo da Vinci

Back in 2017, an oil painting called Salvator Mundi (Savior of the World) sold for $450.3 million at Christie’s auction house in New York. That made it the world’s most expensive by some margin. The painting is one of fewer than 20 thought to be by Leonardo da Vinci, although there is still some dispute over this attribution.

There is also another puzzle. The picture depicts Christ holding a glass orb representing the celestial sphere of the heavens. Such a sphere ought to act like a convex lens, magnifying and inverting the robes behind it. However, Christ’s robes are not inverted or magnified but appear with minimal distortion.

Leonardo was well aware of the way glass refracts light. Indeed, his notebooks are filled with depictions of the way light bounces off and refracts from various objects. And this raises the question of why he drew the orb in this way.

Today, we get an answer thanks to the work of Marco Liang and colleagues at the University of California, Irvine. This group has used computer graphics software to reproduce the scene in three dimensions and then studied how light would be refracted through orbs of different kinds.

After comparing their renderings with the original, they have concluded that the orb is not solid at all. Instead, they show that the painting is a realistic physical representation of a hollow sphere with a radius of 6.8 centimeters but a thickness of just 1.3 millimeters. 

First some background. Inverse rendering is a computer graphics technique originally developed to produce physically realistic renderings of virtual scenes by simulating the physics of light flow. One goal of this technique is to better simulate the appearance of transparent and semi-transparent objects made of glass or water.

The technique begins by creating a 3D representation of the scene, incorporating the texture and structure of all the objects that light interacts with. The scene must also include a source of light and a viewpoint. Then a ray-tracing algorithm maps out the way light illuminates the scene, as seen from the viewpoint.

Liang and co begin by re-creating a virtual version of the painting. “We depict the scene geometry using a rough approximation for the subject’s body along with more detailed representations for the orb and the hand holding it,” they say.

By comparison with the hand, they estimated the diameter of the orb to be 6.8 cm and its distance from the body to be 25 cm. They also refined the geometry of the orb-holding hand to make it touch the orb softly, using Maya, a type of 3D modeling and animation software.

By studying the shadows in the painting, the team concluded that the subject was lit by a strong directional light source from above as well as by a general diffuse illumination. At the same time, they estimated that the viewpoint in the picture is about 90 cm away from the subject.

“With the virtual scene ready, we tested whether the orb was solid by comparing renderings of a solid and a hollow orb,” say Liang and co.

A. Rendering of a solid sphere B. Rendering of a hollow sphere
A. Rendering of a solid sphere B. Rendering of a hollow sphere

The results make for interesting reading. The only way the team is able to reproduce the original painting is with a hollow orb. Furthermore, a hollow orb distorts the background in a specific way. For example, a straight line that passes through the center of the orb is not distorted. By contrast, straight lines that do not pass through the center of the orb are distorted in a way that creates a discontinuity at its edge.

In the painting, Christ’s robes are folded so that five lines appear to pass behind the orb. However, four of the lines have a fan-like arrangement that converges at the orb’s center.  Consequently, there is no discontinuity visible in the reconstructed image or in the original.

However, the fifth fold does not follow this pattern, and the reconstructed image shows a clear discontinuity. The artist blurred this part of the painting where the fold enters the orb. This strongly suggests that he was aware of the way a hollow sphere distorts straight lines that pass behind it.

The team also experimented with varying the hollow orb’s thickness, the results suggesting that it cannot have been thicker than 1.3 mm.

An interesting question is whether Leonardo would have had access to the materials, light sources, and knowledge of optics that the new work suggests he must have had. On the point of optics, Liang and co have studied Leonardo’s notes and think this knowledge must have been well within his grasp. Hollow glass balls were well known at the time and appear in many paintings of the era. And Renaissance artists were experts in re-creating certain lighting conditions.

So Liang and co are sure of their conclusion: “Our experiments show that an optically accurate rendering qualitatively matching that of the painting is indeed possible using materials, light sources, and scientific knowledge available to Leonardo da Vinci circa 1500,” they say.

Of course, the team are not the first to suggest that the orb is hollow—Leonardo’s 2017 biographer Walter Isaacson makes a similar suggestion, and others have discussed it too. However, Liang and co are the first to show that the painting is a physically realistic rendering of a hollow orb and not a solid one.

That will help settle at least some of the controversy over the picture and its whopping price tag.

Ref: arxiv.org/abs/1912.03416 : On the Optical Accuracy of the Salvator Mundi

Deep Dive

Computing

Inside the hunt for new physics at the world’s largest particle collider

The Large Hadron Collider hasn’t seen any new particles since the discovery of the Higgs boson in 2012. Here’s what researchers are trying to do about it.

Why China is betting big on chiplets

By connecting several less-advanced chips into one, Chinese companies could circumvent the sanctions set by the US government.

How Wi-Fi sensing became usable tech

After a decade of obscurity, the technology is being used to track people’s movements.

VR headsets can be hacked with an Inception-style attack

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.