Painting With Fire: A History in GANs

Painting With Fire: A History In GANs by Bård Ionson

“a pioneer of crypto and generative art using his GAN-algorithms to help flesh out small universes across mediums.”
CyborgDAO

“Legend”

Pascal Boyart

“Il fuoco cammina con noi, compagno fedele, nemico inaspettato”
Hackatao

Channeling the fire myths of the world, Prometheus, the Phoenix, the gods of the sun and flame. A story of technology brought to humans; to myths of destruction and rebirth. To be immortal; to be a god; it is always a paradox of fire and technology.

The morality of the fire is always in question but we desire it and fear it. And yet it has changed us, we have learned so much since we first touched fire. We used it to destroy the wood, smelt the metal and build the chip. And now we teach and show the fire what we know; what we create and our reflection is there telling us our faults. The GAN is but a single neuron in a blazing star; a greater fire is coming.

Exploring GAN as an art form

The goal for GAN seems to have been to create a way to ask it for a picture of something and it would draw it. The technology was invented to control and make it as realistic as possible. Bård subverts this to use it chaotically and abstractly; the opposite of what the Ai researchers had in mind.

This project is an exploration of the progression in training of the fire and of the improvements in GANs over time. GAN technologies were developed by researchers and programmers but then utilized by artists in many different forms. This project celebrates both. Inspired by the technology and the artists who have utilized it Bård remixes the code and the imagery into a new form of fire. This history will be seen in the art and in the traits.

Progression of GAN technology

Images and videos are generated from 14 different models one of which is aiCAN (Playform) and different resolutions of DCGAN, Stylegan 1,2 and 3, bigGan, biGAN, AttnGan, pix2pix and more. As the art is revealed the historical origins of the fire will be unfolded. The history of the technology, researchers, developers and the artists will be celebrated in Painting With Fire. Look in the descriptions and the traits for the history of GAN Fire. See the GAN Timeline for reference.

In each art piece minted you will find a historical reference to a paper, code, art piece, developer, researcher or artist who has used GAN or Ai technology. Look for it in the traits and in the art description.

There is a beauty in the noise, the randomness before the resolution into fire.

Training

For this project Bård collected images of fire from photos and videos he had captured, art pieces he had creted before, oscilloscope drawings of fire, from public domain sources and from lovers of his art. Thanks to Kelly LeValley, Shortcut, ezra shibboleth and others for the fire training material.

The images of fire were used to train over 13 different models of 13 different types of GAN technologies from some of the earliest code repositories and progressing to the dawn of the diffusion models. You can see the progression in quality over time and many of the anomalies an artistic interpretation of each class of GAN.

Many of the pieces show the training progression with in a training run. Each round of a GAN training is called an epoch. Early epochs yield an abstract form of fire, distortions and glitches as the generator and descriminator work together to emulate what fire looks like. At times the model collapses and progress is lost for an epoch but returns even more powerfully in the next round of training.

Traits

Technique: This contains the name of the technique the artist has used to generate the image coming out of the model or the way the model was trained. Garbage in garbage out as they say in software development.

Normal Inference: using the generator as originally intended. No special tricks.

Forced Oversize: A feature in DCGAN that allows for the generation of two or more “objects” to be generated into an oversized image. It can generate takes two or more objects of fire and combines them but gets confused because it can’t line up the images correctly. See Layout trait for number of subjects in the oversized image. Some are 100 fires all mixed in one large image. The more subjects per image the larger the resolution.

Latent Spacewalk: This is a technique that works on StyleGAN. The artist choses key frames and then asks the code to transition between these points in the vector space of the model. It moves through the multi dimentional space and outputs frames to create a video.

Training Progression: A series of images curated by the artist from the output as the model is trained over time. Each epoch is used to generate a reference image. The artist created a video of the training process.

Generator code bug: Bård introduced a bug into the generator code that created a glitch in the generation of the layers of the images as it gets sized. It changed the resolution of the output and created a new effect of fire.

Next Frame Prediction: A specific technique to make Pix2Pix generate a prediction of the next frame. After it is trained on one video the artist gives it single frame and it predicts the next frame and based on that frame it predicts the next frame, until the video degenerates into abstraction.

Such a bright and wise mind who plays with curiosity in many different levels
Burcin IO

SIDE NOTE: BARD IS A GENIUS.
ROBNESS

A divine gift to humanity, as it holds the key to its own rebirth and radiant paths onwards.
zaphodok

Artist Tributes

Mixed into the 200 art pieces are a few inspired by other artists using GAN. In some Bård attempted to recreate the training set they used and in others he utilized their outputs to train a new model. These artists include Pindar Van Arman, Robbie Barrat, Shortcut, LizaGrace and Obvious.

The use of a Generative Adversarial Network (GAN) to create images of fire infuses the work with a contemporary relevance, as it represents a frontier where technology meets creativity. GANs, by their nature, involve a dialogical process where two neural networks challenge and refine each other’s outputs, which is metaphorically mirrored in the fiery interplay of colors and textures within each square. This speaks to the historical rebellion against traditional art techniques and the embrace of emergent systems to create art. – Artist Insight a custom GPT art critic created by Bård Ionson

Bård’s First GAN Art from 2018

This selection used Art-DCGAN based on DCGAN trained on oscilloscope art. Bård’s first model and first NFT mints

Neural Dream Map 2018 sold for 8 ETH 2023

Really strong work! You have “inspiration” but this is not inspiration it is something new.
alejandro cartagena

It is a high art done by the only artist that could do it justice, and the aesthetics are just gorgeous.
Pindar Van Arman

Painting with Fire is a collection that showcases some of the finest of his work.
batsoupyum

…he builds images that turn some of the annoying variability infecting traditional GANs into powerful visions that demand attention and introspection and won’t let go
Lawrence Lee

Bard is a pioneer in his field, a master of his technique and a talented artist. With his latest series “Painting with fire” he is depicting the early GAN aesthetic as the very fabric from which our dreams and nightmares are made. Brillant!
Benoit Couty

His production is extensive, and this latest series perfectly reflects his systematic approach to the art practice.
Mattia Cuttini

GAN Timeline

Timeline

1973

Harold Cohen and AARON

GAN – Goodfellow

Ian; Pouget-Abadie, Jean; Mirza, Mehdi; Xu, Bing; Warde-Farley, David; Ozair, Sherjil; Courville, Aaron; Bengio, Yoshua (2014). Generative Adversarial Nets (PDF). Proceedings of the International Conference on Neural Information Processing Systems (NIPS 2014). pp. 2672–2680.

CGAN: Conditional Generative Adversarial Network

Mehdi Mirza Simon Osindero

BiGAN – Goodfellow

Conditional GANs (cGANs) Introduced by Mirza and Osindero in 2014

cGANs represented a significant advancement by allowing the generation of images conditioned on certain inputs, such as class labels. This innovation enabled more controlled and targeted image generation.

Mimi Onuoha “Missing Data”

DCGAN – Radford

Alec; Metz, Luke; Chintala, Soumith . “Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks”. ICLR. S2CID 11758569. a pivotal work in making GAN architectures more stable and robust, particularly for image generation.

DCGAN – Dec 2015 Soumith checks in DCGAN code to GitHub.

Gene Kogan

“AI Mirror”: Kogan is known for his works that explore AI’s role in creativity and society. His “AI Mirror” is an interactive installation that uses GANs to transform reflections in real-time.

CoGAN: Coupled Generative Adversarial Networks Ming-Yu Liu

Oncel Tuzel

Wasserstein GAN (WGAN): Introduced by Martin Arjovsky

Soumith Chintala, Léon Botton in 2017, WGAN addressed training stability issues in GANs by using a different approach to measure the distance between generated and real data distributions. This innovation led to more stable training of GANs and improved quality of generated images.

CycleGAN

enabled unpaired image-to-image translation such as turning horses into zebras or apples into oranges, without needing paired training data. This was a major step forward in practical applications of GANs in image processing and artistic creation. Jun-Yan Zhu, Taesung Park, Phillip Isola, Alexei A. Efros

CAN: Creative Adversarial Networks

Generating “Art” by Learning About Styles and Deviating from Style Norms June 2017 Ahmed Elgammal, Bingchen Liu, Mohamed Elhoseiny, Marian Mazzone We propose a new system for generating art. The system generates art by looking at art and learning about style; and becomes creative by increasing the arousal potential of the generated art by deviating from the learned styles.

Memo Akten Webcam Pix2Pix

Mike Tyka – Portraits of Imaginary People

Memo Akten

An artificial neural network looks out onto the world and tries to make sense of what it is seeing. But it can only see through the filter of what it already knows.

Robbie Barrat forks DCGAN to Art-DCGAN and increases size of model to 128 pixels

Robbie Barrat uploads model weights

added model weights he trained on portraits landscapes and nudes.

Age of the AI Artisans exhibition – Dr. Ahmed Elgammal & AICAN

first AICAN exhibition in LA, based on CAN

SRGAN

Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network. Christian Ledig, Lucas Theis, Ferenc Huszar, Jose Caballero, Andrew Cunningham, ´ Alejandro Acosta, Andrew Aitken, Alykhan Tejani, Johannes Totz, Zehan Wang, Wenzhe Shi

Caroline Sinders “Feminist Data Set”

Pindar Van Arman creates robot and Ai software to paint AI Imagined Faces

AI Imagined Faces by Pindar Van Arman wins Robot Art 2018 Incorporating machine-learning technology, Pindar was able to paint evocative portraits with varying degrees of abstraction.

Artist: Albert (Shady Albert) makes GAN art

Artist: Artist X (Shyam Sreevalsan) creates GAN art

Artonymousartifakt creates GAN art

BigBiGan

Jeff Donahue, Karen Simonyan, Large Scale Adversarial Representation Learning, 2018

ThisPersonDoesNotExist.com Launched by Philip Wang

using NVIDIA’s StyleGAN, this website generated endless streams of hyper-realistic human faces, showcasing the capabilities of GANs to the general public.

David Young Known For notable Ai and GAN art

Nissla Artist utilizing Ai and GAN tools

Artist: Adrien Le Fallher – creator of AI art

Peter Bock creator of AI art

Place of Many (Brandi Kyle) Known For: Artbreeder making GAN art from acrylic fluid painting

WaveFunk (Hans Brouwer) SuperRare Ai Artist

GAN Artist Wavefunk

Dr. Ahmed Elgammal creates Freya

Green Genesis, Tropical Inception, Flora, Psychedelic Wisteria

Obvious: SuperRare Artist Notable Collection: de Belamy Family

Sougwen Chung “Drawing Operations”

Tom White “Perception Engines” A visual overview examining the ability of neural networks to create abstract representations from collections of real world objects.

Ganbreeder is released now known as Artbreeder. is a collaborative

machine learning-based art website. Using the models StyleGAN and BigGAN, the website allows users to generate and modify images of faces, landscapes, and paintings, among other categories. https://github.com/joel-simon/ganbreeder

Playform – based on CAN started operations in 2018

Muse.s Endormie.s

Sold at the Grand Palais on Feb 13th 2018 after being assembled on July 2017 from the children of the cloud thousands of AI-generated portraits of all genders, ages and origins generated in 2016 by ALAgrApHY.

ProGAN: Progressive growing of Generative Adversarial Networks

Tero Karras, Timo Aila, Samuli Laine, Jaakko Lehtinen

HBO Silicon Valley TV series chose an AICAN art work HBO Silicon Valley TV series

chose an AICAN art work to be shown in an episode of the show. Ahmed Elgammal

Sofia Crespo “Neural Zoo”

Crespo is known for her work focusing on natural patterns and biological forms created with neural networks, including GANs

Bård Ionson writes SAGE Anomaly

in words and imges fusing the physical world with fiction and GAN images. A science fiction story that continues to this present time and into the future.

Zaza (Jacob Horne) SuperRare Artist

Notable Collection: Earth Porn

Joy Buolamwini

examining racial and gender bias in facial analysis software

BigGAN – Brock

Andrew; Donahue, Jeff; Simonyan, Karen. Large Scale GAN Training for High Fidelity Natural Image Synthesis. International Conference on Learning Representations 2019. arXiv:1809.11096. BigGAN, High-Resolution Image Generation:.

Obvious

“Edmond de Belamy,” In 2018, a collective named Obvious created “Edmond de Belamy,” an artwork generated by a GAN. This piece was one of the first instances of GAN-generated art being sold at a major auction house (Christie’s), fetching an impressive price and marking a significant moment in the intersection of AI and art.

Next Frame Prediction – Ryosuke Nakajima

Image-to-Image Translation with Conditional Adversarial Networks Phillip Isola

Jun-Yan Zhu, Tinghui Zhou, Alexei A. Efros a significant development in image-to-image translation using conditional GANs. It showed how GANs could be used for tasks like turning sketches into photographs or day scenes into night scenes.

Bård Ionson mints his first art work on SuperRare as an NFT. Alien Intelligence

AI art exhibiting for first time in Art Basel Miami

Dr. Ahmed Elgammal & AICAN at Art Basel Miami

Pierre Huyghe UUmwelt

Lynn Hershman Leeson

“Shadow Stalker”

Kevin Abosch conceptual artist creates Nascent Space

Kevin Abosch creates conceptual GAN art using his own photographs in a series called Nascent Space. Also created Nail(2019) a GAN generation using photos of nails

obxium creates GAN art

AICAN Ahmed Elgammal

first AI solo exhibition in NYC gallery for AICAN, at HG Contemporary

Mario Klingemann

“Memories of Passersby I,” 2019, Sotheby’s Auction:** Klingemann, a pioneer in AI art, used a GAN to create a continuously changing video stream of portraits. This artwork was sold at Sotheby’s, marking another significant sale of GAN-generated art. $52,634

Ai CryptoCoins by Bård Ionson

on SuperRare – art of coins using DCGAN – 22 on SR and 1 on KnownOrigin

Realti Ai art project

Rare Art Festival 2 exhibits GAN art

Bård Ionson exhibited GAN art Air-Coin-Airdrop & The HODL Frame of North Bay Canada & Keven Abosch exhibited Expulsion 16.7 (2017) & Martin Lukas Ostachowski exhibited also.

High-Fidelity Image Generation With Fewer Labels Mario Lucic

Michael Tschannen, Marvin Ritter, Xiaohua Zhai, Olivier Bachem, Sylvain Gelly May 2019 https://arxiv.org/pdf/1903.02271.pdf

SAGAN

Zhang Han; Goodfellow, Ian; Metaxas, Dimitris; Odena, Augustus. “Self-Attention Generative Adversarial Networks”. International Conference on Machine Learning. PMLR: 7354–7363.

GauGAN: Semantic Image Synthesis with SPADE Semantic Image Synthesis with Spatially-Adaptive Normalization

Taesung Park, Ming-Yu Liu, Ting-Chun Wang, Jun-Yan Zhu

“Faces In The Cloud”

Bård Ionson paints an oil painting of a GAN generated image “Faces In The Cloud” and mints it as NFT on Codex.

GAN artist 0xBull GANerative Landscape

Liam Ellul – self portrait using GAN

Mike Almond – GAN art

Artist The Galois Connection mints GAN art An Office Building in Seven Styles

Damion Darko Martin Bammes GAN Art

kromond GAN artist

Bård Ionson Solo Exhibtion ARTificial Creating The Future Out Of History

NVIDIA introduced StyleGAN in 2018

which represented a major advancement in the field of computer-generated imagery. StyleGAN significantly improved the quality and resolution of generated images, leading to hyper-realistic results. This technology became widely known for its ability to generate highly detailed human faces.

Noah GAN artist

Sarah Zucker

Digital artist using analog methods to create art that reflects the past into the future.

Rutger van der Tas created art with GAN

Bloemenveiling – Anna Ridler

Sparrow (Black Box Art) creates with GAN “Lady Liberty Walks Away”

Peter Bock creates “Seas of Europa – Still Frame” with GAN tech

Woman In Boots by solis

Daniel Jakobian

Coming Home For Christmas GAN art by MLIBTY

Germ Cell Cycle GAN art by Uwe Dresemann

DadaGAN Alexander Reben

Kate Crawford and Trevor Paglen

“Excavating AI: The Politics of Training Sets for Machine Learning (September 19, 2019) https://excavating.ai

Anna Ridler

“Mosaic Virus” used a GAN to generate tulip images based on Bitcoin price fluctuations, reflecting on the tulip mania and digital currency.

Trevor Paglan and researcher Kate Crawford’s ImageNet Roulette project

Yuma Kishi creates with GAN

Journey With Ai – artist: illa_2020

Lukas Opekar

ProofOfArtwork (Hans Benzin) collaborates with Bård Ionson

on two Ai music videos using Ai music by Hans and Pix2Pix interpretations of the music by Bård Ionson. Bård again used oscilloscope animations of the music to train a model and generate a video.

Yetepey Creator of The Yetepey Series

WGMeets Creator of ⁠Ganergy

Shortcut

SuperRare Artist and creator of ⁠Shortcuts. Artist used Ganbreeder and Artbreeder

ProofOfArtwork (Hans Benzin) /h3>
Known For: Cryptoportraits and AI music albums in PROOFofSOUND<
Read more

2019

Lukas Opekar (Hoxgen)

Superare Artist & Makersplace AI artist

Alex Shell creates GAN art

Coldie Creator of ⁠Gandinsky

Yura Miron SuperRare Artist creating with Ai

Bitcoin

Creator of many Early AI Art Projects : digital art collective etherislands, ⁠abstract art collection, ⁠Ai human art, ⁠nfttracks

DeepBlack

DeepBlack collective begin minting 3,074 GAN art pieces

Cslasha (Atay Ilgun) Creator of ⁠Realiti

Infinite Skulls AI Paintings By Barrat And Barrot

Painter vs Ai artist. Creating a GAN model from paintings of skulls.

Zebbler creates GAN art for SuperRare

Stina Jones

Creator of ⁠Gandoodle Specimens

Mattia Cuttini

creates GAN art using Playform and his art of cutouts and rubber stamps

Outside Forces Art Enables Gallery “Close Our Eyes” by Bård Ionson

Marjan creates Capital #1 with GAN Generated Paintings

Samogan 2 by ArtNome Jason Bailey

Berk “BASTARD GAN PUNK INCUBATION AUTOMATION UNIT #10947”

Distancer by Anton Dee

StyleGAN 2 – Karras

Tero; Laine, Samuli; Aittala, Miika; Hellsten, Janne; Lehtinen, Jaakko; Aila, Timo (June 2020). “Analyzing and Improving the Image Quality of StyleGAN”. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). IEEE. pp. 8107–8116. arXiv:1912.04958. doi:10.1109/cvpr42600.2020.00813. ISBN 978-1-7281-7168-5. S2CID 209202273

Sven Eberwein – Deep Meme #3

Jascha Suess – Lucid

Bård creates Fountain a 501 piece collection of images created from a StyleGan 2 model of toilets.

Sweep By Tommy

GAN36 [Cultured Meat] – Dave Estes

Scott Eaton “Hyperbolic Composition I – Genesis”

Bård collaborates with Lawrence Lee

a notable artist known for his visions of the American southwest. They created a series of 13 artpieces called Convergence. Bård spent months training GAN models of the 50 years of paintings by Lawrence Lee. Released on SuperRare in 2020

Virtual Rare Art Festival

Shortcut creates Shortcuts

Fountain Remix Bård Ionson – Innovation of GAN toilets and remixing art.

Sofia Crespo & Feileacán McCormick “Specimens”

Bård collaborates with Gisel X

on “Collaboration Of Dreamtime” a GAN study of a journey of a person and a landscape.

Daniel Ambrosi Pestilent Trump

Bård Ionson’s “The Gate”

sells for a record 8.2 Eth on Async.art 24 hour animated GAN art

Bard creates “The Android of Dorian Grey VI”

on AsyncArt a day and night rotating GAN art. Sold for 5 Eth

Bard creates “Dr. Jekyll is Mr. Hyde”

a GAN art that changes from day to night. It is an early example of extending the frame of GAN art so it is not square.

Bård Ionson interviewed on the radio for the show Art Box DMV

Daïm Aggott-Hönsch creates Cressida in Reverie with GAN

Travis LeRoy Southworth uses the Playform AICAN for The Watcher (Clown Cloud-1)

Isabella Salas’ “Small Flower Bits”

Hacking the Tower “Decentralized Culture Lost”

by Bård shown on the D&F Tower in Denver

CADAF 2020 onlne showcased these works by Bård

Global Reality, Artificial Artist (Deep Fake Max), Dreams of Internet Dweller Seven, Latent Spacecraft, The Artificial Intelligence Spawns, SAGE Anomaly Redux, Welcome To Earth, Artificial Intelligence Triality, Maximum TV Ratings

Erin creates “neural triptych” using GANN

AI More Than Human Exhibition

This international exhibition, which traveled to various locations including the Barbican in London, showcased the role of AI in art and included GAN-generated artworks.

Holly Herndon: “CLASSIFIED”

TransGAN

Jiang, Yifan; Chang, Shiyu; Wang, Zhangyang (December 8, 2021). “TransGAN: Two Pure Transformers Can Make One Strong GAN, and That Can Scale Up”. arXiv:2102.07074 [cs.CV].

StyleGAN 3 – Timo

Karras, Tero Aittala, Miika Laine, Samuli Härkönen, Erik Hellsten, Janne Lehtinen, Jaakko Aila (June 23, 2021). Alias-Free Generative Adversarial Networks. OCLC 1269560084.

Jake Elwes: “Zizi – Queering the Dataset”

including those using GANs, have been part of NFT exhibitions and sales. His work typically explores AI ethics, queerness, and the nature of generative algorithms.

Paul Mouginot

of aurèce vettier creates “then came clouds of smoke”

Bard’s The Treachery Of Tokens sells for 11 Eth

Adrià Gil

creates “the spanish civil war bombs” with GAN technology

Crossing The Interface

by Holly Herndon and Mat Dryhurst

VQ-VAE-2 for High Fidelity Image Synthesis

Ali Razavi, Aaron van den Oord, Oriol Vinyals The Vector Quantized Variational Autoencoder (VQ-VAE-2) showed impressive results in generating high-fidelity images, pushing the envelope in terms of image quality and resolution in GANs.

Scott Eaton

“Entangled”: Eaton’s work combines classical artistic techniques with advanced computer algorithms, including GANs, to create sculptures and other artworks.

Bard’s Freaky Faces

are made from faces made with SPADE GauGAN and some are animated using thin plate spine motion model. The faces were constructed from the artists drawings of faces with extra eyes noses and mouths.

LizaGrace mints “Colorform Metaverse”

CryptoArt Revolution Paris Exhibition

“Knights of the Apocalypse”, “Color of Crypto by Bård Ionson

Outside Forces Art Enables Gallery “Presidential Portrait for 2017- 2021”

by Bård Ionson

Abstract Art in the Age of New Media

Museum Of Contemporary Digital Art “Life Is Green”, “Alien Life of Venus” Bård Ionson

Linda Dounia “Dust is hard to breath” based on 2

000 images based on her own acrylic paintings and inspired by the Fremen in Dune.

Genesis

Clair Silver

Pindar Van Arman creates a Quantum Portrait of Bård Ionson

Bård Ionson Announces a Limited Edition Sage Anomaly Series Release: The Entity

Distortion Genius: Bård Ionson

a Singular Artist for a New Age written by William Peaster

Multiple pieces of art by Bård Ionson

are celebrated at Historic Art Fest

Bård Ionson interviewed for the second time on ArtBox DMV radio show

“The Dance (of Fire)” by Bård Ionson

exhibit at Ça sent le Crypto-Sapin Paris NFT Factory

CADAF 2022 New York City

Bård Ionson exhibits and sells multiple GAN art pieces. Soul Scroll interactive blockchain prayer machine is also exhibited for the first time.

Historic NFT Fest Barcelona

“Representation Of The Entity” by Bård Ionson

“HODL Frame of North Bay Canada”

at Icons of CryptoArt Dubai

“PodGans” Pindar Van Arman

Pindar Van Arman

creates quantum Ai art “Quantum Portraits” Did he use a GAN?

VQGAN-CLIP: Open Domain Image Generation and Editing with Natural Language Guidance

Katherine Crowson, Stella Biderman, Daniel Kornis, Dashiell Stander, Eric Hallahan, Louis Castricato, Edward Raff – innovation in prompting a model with language to produce images

Crypto Art Begins

book is released featuring art of 50 cryptoartists including Bård Ionson

Anne Spalter – Ai Spaceships (Night Cafe)

Anne Spalter is an established digital mixed-media artist known for her innovative use of technology including artificial intelligence and GANs (Generative Adversarial Networks), in creating art

Pascal Boyart – “Legend”

Mattia Cuttini “Back And Forth”

exhibition NFTFactory Paris contains multiple GAN created pieces.

Kill Them All”

For the second time Bård Ionson and Lawrence Lee collaborate on “Kill Them All” made from GAN and hand drawings curated by Eleonora Brizi of Breezy Gallery

CyborgDAO

“Ionson is one of the first crypto artists to mint on the blockchain in 2018. With an algorithmic style which favors cohesion of color where edges blend objects meld together while subjects always remain recognizable in both political and crypto-specific artwork.”

IHAM Gallery MoCA Collection “Color of Crypto”

“Bard’s Freaky Faces” by Bård Ionson

Bård Ionson’s work called “Battledore” is donated to the Oxford Bodleian Libraries c

ollection and appears in the Alphabets Alive Exhibition. The images from this work were used to train Bårds first GAN model used for many early pieces.

“Deep Fake Max (Artificial Artist)” donated to the Pompidou Museum Paris

Bård’s art “Deep Fake Max (Artificial Artist)” donated by Dan Polko to the Pompidou Museum Paris

Collector: Burcin burcin.eth

“From the day zero I figured out that there s a digital art communityBard’s style and approach caught my attention. I’m big fan of experimenting and it’s guaranteed to find it in every aspect of Bard Ionson art as crafted masterfully. Such a bright and wise mind who plays with curiosity in many different levels”

Bård is slected for the NFT Art Prize 2023 by Arab Bank Switzerland for ‘Future Is Where East & West Touch” not GAN art but made with Stable Diffusion

La Trahison de l’IA (The Treachery of Ai)

100% created by artificial intelligence – or not exhibition NFT Factory selected Bård for an Ai art exhibition in 2023. Bård produced “La Trahison de l’IA (The Treachery of Ai)” A self portrait of his Ai self speaking French.

Painting With Fire: A History in GANs

sovrn curated

OpenSea

Rarible

Sansa

Exploring GAN as an art form

Progression of GAN technology

Training

Traits

Artist Tributes

Bård’s First GAN Art from 2018

GAN Timeline

1973

Harold Cohen and AARON

2014

GAN – Goodfellow

2014

CGAN: Conditional Generative Adversarial Network

2014

BiGAN – Goodfellow

Nov 2014

Conditional GANs (cGANs) Introduced by Mirza and Osindero in 2014

2015-2022

Mimi Onuoha “Missing Data”

Nov 2015

DCGAN – Radford

Dec 2015

DCGAN – Dec 2015 Soumith checks in DCGAN code to GitHub.

2016

Gene Kogan

2016

CoGAN: Coupled Generative Adversarial Networks Ming-Yu Liu

Jan 2017

Wasserstein GAN (WGAN): Introduced by Martin Arjovsky

March 2017

CycleGAN

June 2017

CAN: Creative Adversarial Networks

May 2017

Memo Akten Webcam Pix2Pix

2017

Mike Tyka – Portraits of Imaginary People

2017

Memo Akten

Aug 8 2017

Robbie Barrat forks DCGAN to Art-DCGAN and increases size of model to 128 pixels

Aug 15, 2017

Robbie Barrat uploads model weights

October 2017

Age of the AI Artisans exhibition – Dr. Ahmed Elgammal & AICAN

2017

SRGAN

2017-current

Caroline Sinders “Feminist Data Set”

2018

Pindar Van Arman creates robot and Ai software to paint AI Imagined Faces

2018

Artist: Albert (Shady Albert) makes GAN art

2018

Artist: Artist X (Shyam Sreevalsan) creates GAN art

2018

Artonymousartifakt creates GAN art

2018

BigBiGan

2018

ThisPersonDoesNotExist.com Launched by Philip Wang

2018

David Young Known For notable Ai and GAN art

2018

Nissla Artist utilizing Ai and GAN tools

2018

Artist: Adrien Le Fallher – creator of AI art

2018

Peter Bock creator of AI art

2018

Place of Many (Brandi Kyle) Known For: Artbreeder making GAN art from acrylic fluid painting

2018

WaveFunk (Hans Brouwer) SuperRare Ai Artist

2018

GAN Artist Wavefunk