In May, we took a day trip to Linz, a UNESCO City of Media Arts, to check out the ARS Electronica Center, an institute and creative ecosystem combining the fields of art, science, and technology. In this blog entry, I will cover some projects that caught my attention.
Biology and Art
One of my highlights was seeing the research and various approaches stemming from biology.
For example, Interwoven, an artwork by visual artist Diana Scherer. By studying plant root growth, she developed a biotechnique in the form of various templates that the root system adapts to. The result shown was an intricate, textile-like pattern. This is why I was drawn to this research, as the resulting patterns were a mix of human intervention and biological adaptivity, creating various interesting textures and shapes with naturally occurring randomness rather than digital computation.
One more textile design practice I was drawn to was Growing Patterns, Living Pigments by Julia Moser, which explores dyeing processes via pigment-producing bacteria in combination with various techniques. The bacteria used, which naturally produces a bluish-purple pigment called violacein, reveals organic patterns and hues, as well as showcase a potentially sustainable approach to textile coloration.
Both of these works are based on co-creation with non-human biology. Organoids by Jürgen Knoblich and Madeline Lancaster, in a way flips the script by using human stem cells as the starting point. Created for scientific research rather than artistic production, the project focuses on organoids, masses of cells grown from stem cells with brain-like structures similar to the embryonic brain. The project gives new insights into how the human brain is formed, which can help researchers understand the development of brain disorders.
~21,600 by Dorotea Dolinšek looks at the human through the lens of an organism defined by an intrinsic, vital dependence on air. The resulting version shown at ARS was a prototype of the work, made by blowing coal dust onto paper over the course of 9 hours, using her breath as the central gesture. This work interested me due to its results: a collection of coal dust dots, none exactly the same, creating a pattern that could be interpreted as an algorithm and potentially used for sonification and algorithmic composition. In the end, she made it into a video performance spanning 24 hours, featuring a mask with a microphone as the input and an airbrush attached to the nose of the mask as the output.
AI and Art
During our visit, we got a tour of the Piano Room, equipped with a Yamaha DC1, a grand piano capable of self-playing, attached to an AI-trained algorithm for visualization developed at Ars Electronica Futurelab. For a practical demonstration, I got to play an improvisation. It was intriguing to see the resulting visuals, as the algorithm was trained mostly on classical music. While playing a repeating phrase, it visually adapted and formed a sort of memory of the visual pattern, referencing it for a short moment afterward.
While in no way new, it was still interesting to see and interact with installations using GANs (generative adversarial networks), a generative AI framework comprised of two neural networks that compete against one another to create synthetic data.
Pix2Pix: GANgadse lets users draw sketches, which are turned into cat pictures. This was done by feeding the algorithm cat images and training it to recognize typical features, like pointy ears and cat-shaped eyes. In practice, it reads your drawn shapes and approximates a guess. This was interesting to play around with, as drawing nonsensical sketches produced abstract and slightly unsettling flesh-like objects. We also tried drawing an elephant, which it turned into a fuzzy creature, and a cat, which worked as intended.
ShadowGAN uses a camera to capture the outlines of visitors and, using the algorithm, reinterprets those silhouettes as landscapes. In this case, the AI was trained with images of mountains and their landscapes, with the aim of interpreting everything it sees through this lens. The result was an interactive installation comprised of various abstract landscapes.
Both of these projects create digital forms that adapt to incoming input, almost like digital organisms.
3D 8K Experience
Lastly, I’d like to mention Deep Space, a 3D experience space developed by the Ars Electronica Futurelab. The room consists of eight 4K projectors, able to produce 8K when paired in twos, projected onto the wall and floor to create one continuous space. This is combined with 3D shutter glasses for visitors and a laser tracking system, though in our case, the interactive tracking capability wasn’t put to full use, since we experienced the program seated rather than moving through the space ourselves. The space also features a modular multichannel audio system.
Our guided tour through the experience consisted of nautical and space themes, with the most memorable moments being a 3D projection of a whale, followed by an artistic piece consisting of 3D-projected lines engulfing the audience that act as a translation and visualization of a musical composition.
Sources:
https://ars.electronica.art/news/en/
https://ars.electronica.art/futurelab/en/projects-deep-space-evolution/
https://dianascherer.nl/
https://www.livingpigments.com/
https://ars.electronica.art/center/en/organoids/
https://www.doroteadolinsek.com/projects/21-600-2
https://ars.electronica.art/center/en/piano-room-2/
https://en.wikipedia.org/wiki/Generative_adversarial_network
https://ars.electronica.art/futurelab/en/projects-understanding-ai/
https://ars.electronica.art/center/en/pix2pix-gangadse/
https://ars.electronica.art/solutions/en/deepspace/
https://ars.electronica.art/futurelab/en/projects-deep-space-evolution/
