AI Trained to Generate BAYC-Style Images

A machine-learning expert, Yannic Kilcher, has created an application that генерирует images in the style of the NFT collection Bored Ape Yacht Club (BAYC).

The system is based on Generative Adversarial Networks (GAN). Kilcher studied the BAYC collection of 10,000 images and trained the NVIDIA StyleGAN2 algorithm on the assembled dataset.

The finished system produces “boring monkeys,” which are completely unique and interchangeable, the developer said.

However Kilcher acknowledged that the images do not always come out perfectly.

“The quality of my system is not very good, artifacts are often visible,” he said.

The developer added that his colleagues Cyril Zakka and Nathan Cooper Jones created similar tools that yield better results.

You’ve probably heard of exclusive #NFTs like the Bored Ape Yacht Club that sell for thousands of dollars… I had a bit of free time this week so I trained a small AI model to generate entirely new ones with a bit of help from @PDillis and @karlhabashy. pic.twitter.com/eD2L8rVPwu

— Cyril Zakka (@cyrilzakka) November 25, 2021

In April, an unknown breached the Instagram account of the Bored Ape Yacht Club project.

Earlier, OpenAI unveiled the second version of its image generator for text descriptions, DALL-E.

Subscribe to ForkLog AI news on Telegram: ForkLog AI — all the news from the world of AI!