Pablo Picasso, like many geniuses of the art world, went through difficult times at the beginning of his career. Being quite poor, he could barely afford painting materials and canvases and sometimes had to paint over previous works to recycle the canvases. Some of these covered paintings have been identified using various imaging methods such as X-ray, UV and infrared, but for the most part they have only been able to identify the edges of the lost paintings.
Recently, with the help of artificial intelligence, it was possible to fill the gaps, and with the help of 3D printing, a team of specialists managed to reproduce the lost painting without damaging the priceless work of art that covers it.
Research group – Oxia Palus – specializes in recovering lost works of art using artificial intelligence and other technologies. The company recognizes that there may be thousands of lost masterpieces hidden beneath other paintings, and this method could help the world see what was once lost.
The blind man’s breakfast
Picasso’s paintings from this period of poverty became known as the Blue Period, and one such painting from that era is The Blind Man’s Breakfast (painted in 1903), which you can see in the image below.
The existence of the painting hidden underneath The blind man’s breakfast was discovered using X-ray fluorescence (XRF) imaging in 2010. This hidden image became known as The solitary crouching nude and you can see what it looks like below. The researchers first scanned The blind man’s breakfast with X-ray and infrared images to determine the shape and characteristics of the hidden painting. Then, an artificial neural network was programmed to reproduce characteristics of Picasso’s work, such as brushstrokes.
Texture was added by generating a height map of the image, and the AI-generated brush strokes were then replicated at 1:1 scale on a canvas with a 3D printer. Short, 3D brush strokes give three-dimensional characteristics and are far superior to direct 2D color printing.
Which makes sense.
You can see the result below, which is a combination of scans and brushstrokes, based on Picasso’s style of the time.

The hidden painting has long been recognized as part of the blue period of the painter’s work, and with the help of the 3D printed artwork, it will be the first time the public will be able to see the hidden work.
“I hope Picasso is happy to know that the work he had hidden from future generations is finally being revealed, 48 years after his death and 118 years after the painting was hidden,” said George Cann, co-founder of Oxia Palus
🎨 The Challenge of Uncovering Hidden Art
1. The Superimposed Paint Layers
When Picasso painted over a previous work, he usually applied a new, opaque layer of paint (often a ground layer) to hide the original image.
-
X-ray Radiography: X-rays are often the first tool used. They can easily penetrate the paint layers. However, they only reveal the presence of heavy elements (like lead white pigment, or certain elements in cadmium yellow or cobalt blue) used in the underlying and upper layers. If the underlying painting and the visible painting used similar heavy pigments, the X-ray image becomes a confusing superposition of both images, making it difficult to distinguish the hidden composition’s details—often, only the strongest lines or edges stand out.
-
Infrared Reflectography (IRR): Infrared light is highly effective at penetrating most paint layers but is then absorbed by carbon-based materials (like black chalk underdrawings or charcoal). While IRR can reveal the artist’s initial sketches beneath the top paint layer, it may not clearly show the underlying painted composition if the paint layers themselves are too thick or contain interfering pigments.
-
Ultraviolet (UV) Fluorescence: UV light reveals details about the surface, such as varnishes, restorations, and different types of binding media, but it rarely penetrates deep enough to show the full composition of a hidden painting.
2. The Solution: Advanced Scanning Technology
To get beyond just the edges and fully resolve the composition of the hidden painting, researchers are now using highly sophisticated, targeted imaging techniques:
-
Macro X-ray Fluorescence (MA-XRF) Scanning: This is the game-changer. MA-XRF scans the painting point by point, identifying the elemental signature of the pigments in each layer.
-
For example, if the hidden painting used a lot of chromium (found in certain greens) and the top painting did not, researchers can create a digital map showing only the areas containing chromium, effectively «seeing through» the top layer to isolate the hidden painting’s composition.
-
This technique produces separate, high-resolution images for each element present, allowing art historians to virtually reconstruct the lost work, which is why they are often used to confirm the identity of the hidden works.
-
-
Neutron Activation Autoradiography (NAA): An older, highly complex technique used to analyze the elemental composition of the paint layers by bombarding the painting with neutrons, but it is less common now due to the complexity and safety risks involved.
By using MA-XRF and advanced digital processing, art historians can often move beyond seeing just the edges or ghost image and reveal the full subject of Picasso’s lost masterpieces.
latest posts published
What can 3D scanning do for reverse engineering?
Ultimaker opens new possibilities for industrial printing!
Fields of applicability of 3D technology
3D scanning improves inspection and maintenance of race cars
ABS filaments for 3D printing
PET filaments for the 3D printer
How to optimize mold inspection for auto parts
The best large format 3D printers. The buyer’s guide
How architectural models are created using additive manufacturing
