September 30, 2022
  • September 30, 2022

Computer-generated sculptures are eerily real

By on August 3, 2021 0




Matteo Rattini, “This sculpture does not exist (seed-5117, seed-6832 and seed-3309)” (2021), digital images (© Matteo Rattini)

Where have you seen this sculpture? Maybe at an art fair, in one of the hundreds of dizzying stalls, or maybe in the glossy pages of a design magazine? The answer is neither, as none of the works pictured above are real. If the sculptures seem strangely familiar, it’s because they are.

For his project “This sculpture does not exist“, artist Matteo Rattini trained a neural network – a set of algorithms designed to recognize patterns – to create images of contemporary sculptures based on Instagram’s suggestions. Totemic structures, biomorphic shapes and forms The resulting sleek stackings against drab gray backgrounds have an eerie mass-produced patina, also vaguely similar to Ikea furniture.

“The algorithm simply adapts to user tastes, creating an artificial environment in which diversity is replaced by repetition and standardization,” Rattini told Hyperallergic. “While seeing footage from a show you haven’t been to yet might spoil the experience, the algorithms and strategies used by social media in managing content work on a deeper level, altering and reshaping the perception of art itself. “

Rattini, who is currently studying multimedia arts at Università Iuav di Venezia, wanted to understand how Instagram’s algorithm impacted his experience of art as well as his own practice.

“The way to find out was to expose the neural network to the same amount of images and information and see what it would produce,” he said.

Matteo Rattini, “This sculpture does not exist (seed-3539)” (2021), digital image (© Matteo Rattini)

The artist opened a new Instagram account, followed a few contemporary art profiles, and created a script that would automatically engage with suggested posts – what he calls “biting into algorithm recommendations.”

At first, Instagram’s suggestions were broad and diverse. “But over time everything started to look the same,” Rattini said. “From images of paintings, installations, performances, he slowly reduced to images of minimal modernist sculptures presented inside an aseptic white cube.”

The artist then fed around 4,000 photos of artwork into a Generative Adversarial Network (GAN), a type of neural network made up of two competing algorithms – a discriminator, which evaluates images in a dataset, and a generator, which produces random images. Based on the discriminator feedback, the generator slowly improves, producing increasingly realistic visuals.

The “dataset” of real sculptures used by the algorithm to generate digital images. (courtesy Matteo Rattini)

“In a way, this mechanism reflects how our brain works, in a very simple way: you give it a lot of information about trees, for example, and when it comes time to draw a tree, not just your idea of ​​a tree informed by all the trees you have seen, but as you draw it your brain is constantly judging it through your eyes and giving real-time information on how to compensate and adjust, ”he said. Rattini explained.

Indeed, computer-generated sculptures have become increasingly compelling, evolving from blurry and amorphous spots to clearly defined, structured and realistic imitations of art. Any apparent diversity in form, texture and color is flattened by a more pernicious similarity, an underlying monotony more difficult to pin down or describe.

The generator images were getting more and more convincing. (Left: A work in progress; right: “This sculpture does not exist (seed-4330 ″) (2021) (© Matteo Rattini)

The project helped Rattini visualize the otherwise imperceptible processes implemented by platforms like Instagram to tailor their users’ experiences, with the inevitable result of homogenizing the content they see – showing them works that they are assured of. to love, or, more dangerously, political opinions that they already agree with.

“Every post you like, every image you share, every topic that interests you becomes data used to train recommendation algorithms to better understand you, manipulate you and predict what you’ll like and what will keep you online.” , Rattini said.

Still, Rattini says he’s not against social media; he just wants to get a better idea of ​​how it works and how it might affect us.

“For me, it comes down to always remembering that the same tool that made me discover artists like Rothko or Bacon could alter my perception of art and my artistic production if it is not used consciously”, he declared.

Matteo Rattini, “This sculpture does not exist (seed-3539)” (2021), digital image (© Matteo Rattini)

The act of touching allows for a deeper sensory understanding for the viewer while simultaneously creating a rebellion against the terms of visualization, the terms defining the museum and gallery space.


Photographer Fin Serck-Hanssen follows Hedda, a Norwegian woman in her twenties, as she travels for cosmetic surgeries and vaginoplasty.


Sean Baker’s flick about a stranded pornstar seducing a teenage girl is sexually blunt like few American films are today.