Artificial Intelligence Art

Exploring computer generated art

[artificialintelligenceart]

Summary

Undoubtedly in the last two decades scientists and researchers have achieved outstanding progress in technology while artificial intelligence is one of the hottest topics lately. AI is not only used in every industry nowadays but it seems that it is capable to replicate actions that were once thought to be unique for humans, like logic, reasoning, perception and creativity. Some people fear out or reject the increasing development in the field while others are fascinated and remain eager to follow every single step of its development.

During our research for this blog we explored in what way artificial intelligence has been implemented in different kinds of digital art. Throughout our posts we saw some works of art where artificial intelligence and digital art are combined with other kinds of art, like music performance or interactive installations. We also had the chance to watch its history, present different aspects of view and see some modern representative artworks which use different AI algorithms or techniques. But it is a really wide field so it wasn’t possible to dig into further details due to the short period of time.

The truth is that AI isn’t something new since it has been used for a long time even in arts. Maybe not everyone knows that most of the modern digital artistic tools have integrated it owing the development and precision of many tools to AI. Since digital art is using digital tools to create art, we could safely support that AI has its own place in the history of digital art.

*The photo is an artwork created by Helena Sarin with AI

[audiovisualartists]

Το INORI – PRAYER* αποτελεί ένα έργο βαθύτατα μελετημένο, που για την υλοποίησή του απαιτήθηκαν τρεις μήνες συνεργασία πολλών ατόμων στην προσπάθεια τους να παρουσιάσουν κάποιες πτυχές της κινέζικής και της ιαπωνικής κουλτούρας μέσα από μια πρωτότυπη καλλιτεχνική έκφραση. Το σύστημα χαρτογράφησης προσώπων επέτρεψε την παρακολούθηση έντονων παραστάσεων, κάτι που ήταν αδύνατο μέχρι τώρα, χάρη στη χρήση του προβολέα DynaFlash που διαθέτει τελευταίας τεχνολογίας και της υψηλής ανίχνευσης ταχύτητας (1000fps).

Είναι ενδιαφέρον κανείς να ερευνήσει την σημειολογική προσέγγιση αυτού του βίντεο, που η προβαλλόμενη εικόνα μοιάζει να είναι ενσωματωμένη σε μέρος του δέρματος και των εκφράσεων ενός ανθρώπινου προσώπου, το οποίο παραμορφώνεται ή μεταμορφώνεται και ενισχύεται εκθετικά. Το θέμα του έργου είναι η ίδια η Ζωή και συνδυάστηκε με το μουσικό κομμάτι που “ενσαρκώνει” τη ραδιενέργεια, με σκοπό να αποδοθούν τα δεινά της τελευταίας που είναι ο θάνατος και παράλληλα η ταλαιπωρία και η θλίψη. Αυτό αποτυπώνεται εικονικά με μαύρα δάκρυα, κρανία, κομμένα πρόσωπα, μάσκες Noh και το Heart Sutra.

Keep reading

[artificialintelligenceart]

A really interesting work which combines for the making artificial intelligence's technique "Real Time Face Tracking" with projection mapping and state-of-the-art technology that creates an impressive result.

[artificialintelligenceart]

The portrait of Edmond De Belamy

The above portrait was made by a GAN network and it is about a fictional person which has never existed. Instead of the artist's signature it has the mathematical concept that created it. Nevertheless, this portrait blew away the art world when it was sold for the extraordinary amount of $432.500 at Christie's auction's house in New York surpassing its estimated value of about $10.000.

Source

After this a lot of controversy was raised around the role of the artist, what is really art and the credits about the work, since the creation of this portrait was based on an open-source network hosted on Github where its developer had already published some of the results he had gotten after training the model.

This has created unease for some members of the AI art community, which is open and collaborative and taking its first steps into mainstream attention. Seeing an AI portrait on sale at Christie’s is a milestone that elevates the entire community, but has this event been hijacked by outsiders?

Source

Art critic Jonathan Jones refuses to acknowledge Edmond de Belamy – or anything created by artificial intelligence – as art at all. He writes that “computers would need to replicate human consciousness before they could replicate the funny thing humans do called ‘art’. Art is a way in which human consciousness expresses itself, and is equally true of the earliest cave art, Rembrandt’s portraits and Duchamp’s urinal. And that is what is missing from Portrait of Edmond Belamy. Art is a way humans communicate ideas, perceptions and feelings to each other.”

Source

[artificialintelligenceart]

Ethical AI

The ethics of AI is a hot topic right now, and for good reason: AI has the power to transform our world. How can we ensure AI remains aligned with human values?

Source

AI artists is the largest community of creative people that have incorporated AI in their work. Apart from using AI they would like to discuss about the way we interact with technology and to ensure that it is being used in a trustworthy and ethical way. On their website they have listed a collection of resources about Ethical AI.

*Photo by Shane Rounce on Unsplash

[artificialintelligenceart]

Quotes

Here are some quotes about artificial intelligence with different or contradicting opinions.

[artificialintelligenceart]

History of Artificial Intelligence

The history of artificial intelligence seems to be very long with its beginning dating back to Greek mythology. As we can see in the Timeline of Artificial Intelligence on Wikipedia, some Greek myths had incorporated the idea of artificial beings and intelligent robots such as Talos who is depicted in the above photo of a didrachma from Phaistos (ca. 300/280-270 BCE). Sacred mechanical statues in Egypt and Greece were thought to be capable of wisdom and emotion. Philosophers were talking about mechanical thought and engineers were creating mechanical men and other automata. A muslim polymath of the middle ages was combining engineering and arts.

Discussions, thoughts and inventions have not ended until today when the development of AI seems to be so rapid than never before. Although Alan Turing was the first to talk about "Computing Machinery and Intelligence" some other scientists had contributed to the development of the field before him. During the twentieth century AI went through some difficult periods in such a way that some scientists started giving different names to their researches in order to avoid negative criticism or to procure funding. Nevertheless, AI had solved some difficult problems and algorithms developed by it had began to appear as parts of larger systems. But the field of AI received little or no credit for these successes as the greatest innovations appeared to be as another item of computer science. As Nick Bostrom explains: "A lot of cutting edge AI has filtered into general applications, often without being called AI because once something becomes useful enough and common enough it's not labeled AI anymore."

Nowadays, the vast collection of the amount of data which has been gathered, mostly due to the use of the internet and the social media, has brought AI to the front stage since all these data need to be processed to give meaningful information. But it has also given artists another interesting but rather complicated tool for inspiration and creation.

Source

[artificialintelligenceart]

An infographic about the History of AI

By Karl Tate on LiveScience

The above infographic was made in 2014 and outlines some of the most significant moments in the history of AI as they are depicted in science and fiction from 1950 to 2014.

[artificialintelligenceart]

Break Free

By Taryn Southern

The instrumentation of the song was the first solo album which was composed with AI while the lyrics and vocal melodies were written by the artist. As we can notice AI has also been used in the making of the video clip.

Taryn Southern is an artist and futurist who has spent a long time working at the intersection of emerging technology and storytelling. Having worked on AR/VR she received a Google grant to create experimental VR series and she was named one of the Top 20 Women in VR. She is a featured AI artist who has performed in some of the most important tech summits.

Source See more

[artificialintelligenceart]

DATAMONOLITH_AI

An Architectural Data Sculpture Installation by Ouchhh in Mori Museum Tokyo.

By using GAN technology and other AI algorithms Ouchhh creative studio created a Data monolith with data from Göbeklitepe in Turkey which are dating back to the Pre-Pottery Neolithic B era.

By trying to understand the human prehistory and civilization they are getting a glimpse of the future and the Arts wondering about our modern civilization and the impact of the advances in technology in our world.

[artificialintelligenceart]

AI Virtual Dance Partner

The LuminAI project from the Georgia Institute of Technology was showcased in 2016 within a dance and technology performance at the Hambidge Art Auction in partnership with the Goat Farm Arts Center. The project was housed inside a 15-foot-tall geodesic dome which was designed and constructed by Jessica Anderson, a digital media master’s student at Georgia Tech. The dome was lined with custom-made projection panels for projection mapping. Its surfaces allow people to watch their own avatar as it dances with a virtual character which learns how to dance and improvises by watching the other person's moves.

Brian Magerko, the digital media associate professor who led the project believes that: “Co-creative artificial intelligence, or using AI as a creative collaborator, is rare. As computers become more ubiquitous, we must understand how they can co-exist with humans. Part of that is creating things together.”

Source

See here for more

[artificialintelligenceart]

Dynamic Texture Synthesis

Inspired by recent work on image style transfer, Matthew Tesfaldet, Marcus A. Brubaker, and Konstantinos G. Derpanis introduced their "Two-stream convolutional networks for dynamic texture synthesis" in IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018. The result is impressive and really very interesting!

Source

[artificialintelligenceart]

Latent Space Interpolation (2019)

By Mattie Tesfaldet

Fot this work, which was accepted at the Computer Vision Art Gallery, Mattie Tesfaldet used a Compositional Pattern Producing Network extended with a novel approach by using Fourier coefficients and an Inverse Discrete Fourier Transform. This way their Fourier-CPPNs provide better visuals for image synthesis.

Source

[artificialintelligenceart]

Artistic Style Transfer Algorithm

Here are some of my experiments with a PyTorch implementation of the "Neural Algorithm of Artistic Style" method which was developed by Leon Gatys et al. For the above output images, I used my own photos as content and some free images from Unsplash for style.

Artistic stylisation of images is a subject that has been studied in computer graphics under the label of non-photorealistic rendering. The difference between these approaches and Gatys' work is that they provide specialised algorithms in order to render a source image in one specific style. Source

[artificialintelligenceart]

Inception Animation with Deep Dream

Deep Dream is a computer vision software developed by Google engineers. It is based on deep learning and specifically uses a Convolutional Neural Network (CNN) to find patterns on images and enhance them creating a dream-like appearance on over-processed images. The software was based on a deep convolutional network which was given the code name "Inception" after the well known movie directed by Christopher Nolan. After the software was open sourced and publicly released by Google, many similar applications appeared giving users the chance to transform images or videos. As it is stated on the website Deep Dream Generator:

"Initially it was invented to help scientists and engineers to see what a deep neural network is seeing when it is looking in a given image. Later the algorithm has become a new form of psychedelic and abstract art."

Nevertheless, creating art with software is a new approach which has raised some questions about the credits for the artwork and if this is indeed a kind of art.

So, you can give it a try for free with your own images to see yourself if you find it inspiring and artistically interesting. If you are more into coding, you can try playing with the original code here.

Read more about its development history and criticism.

[artificialintelligenceart]

"AI, the foundation of all video games"

If you have ever played a video game, you have interacted with artificial intelligence (AI). Regardless of whether you prefer race-car games like Need for Speed, strategy games like Civilization, or shooting games like Counter Strike, you will always find elements controlled by AI. AIs are often behind the characters you typically don’t pay much attention to, such as enemy creeps, neutral merchants, or even animals. But how does AI found in gaming relate to the AI that tech giants talk about every day?

Read more

The above article was written by Dr. Harving Lou as a part of the "Special Edition: Artificial Intelligence" on Harvard's blog SITN Science in the news. Throughout the article Dr. Harving thorougly explains the way AI is used on video games and how players interact with it. He also points out some pertinent questions that arise from the rapid development of modern gaming technologies that widely use AI.

[artificialintelligenceart]

Optics

By Maxim Zhestkov

An amazing experimental art film by Maxim Zhestkov exploring the behavior of artificial light and color in digital environments. The film was created after a series of experiments with programmers and artists trying to achieve different effects and behaviors of light while working with refractive or reflective surfaces. An architectural gallery space is being blended with computer-generated digital glass sculptures, producing animated refractions, chromatic aberrations, colourful reflections and other conditions which are impossible in real life.

Source

[artificialintelligenceart]

"SAY_SUPERSTRINGS"

A Real-Time Generative Installation Performance Of Electrical Activity In The Musician’s Brain! dastrio x Ouchhh x Ars Electronica

We created a cognitive performance that we generate real-time brain waves at the Deep Space with the South Korean/German trio. At a concert which we visualize the changes in brain wave activities in real-time, we will also transform the Delta, Theta, Alpha, Beta and Gamma brain waves into a real-time concert experience that is wrapped around by generating the data about emotion, focus & attention, some auditory and neural mechanisms with electroencephalogram (EEG). What if matter is nothing but notes coming out of a vibrating string?

According to superstring theory, all matter in the world is made up of one thing: vibrating thin strings. These strings, which vibrate at different resonances, bring everything into existence in the known universe. Matter consists of small strings. When these strings are pulled in a certain way – just like a violin or guitar string – they create a frequency. Therefore, the pitches/notes occur. We have become aware of the existence of the little notes that these superstrings create, and we realize that the universe is a symphony, and all the physical laws of the universe fit these superstrings.

Ouchhh will take inspiration from the notes that exist in the universe while micro-strings vibrate (Subatomic Particles) in real time and define the melodies created by the notes as “Matter” and symphonies of these melodies as “Universe”. With dastrio, Ouchhh will take 11 dimensions in abstract directions in super grade gravity theory and move them beyond space in real time. The dimensions captured intuitively in living space will constantly change and turn into reality.

Real time visualization of brain electrical activity has existed for nearly half a century. Early visualization systems required extremely expensive equipment and timely application supports. The present microcomputer revolution provides extremely advanced performance and has helped for the development of better visualization environments. Real-time visualization can improve understanding of more complex phenomena in neuroscience, such as brain activity, while provides assessment of genuine brain dynamics, and perception of inherent spatio-temporal patterns of brain electrical activity. For the development of environments like these advanced computing methodologies have been used including artificial intelligence and computer vision.

Source See more