HENI is delighted to present Color Rhythms, a generative NFT collection by the internationally renowned graffiti artist MadC. Consisting of 1,000 unique NFTs, Color Rhythms innovatively combines a generative algorithm, machine learning and hand curation in order to push the boundaries of MadC’s expressive compositions.
This drop is also novel in its use of machine learning to create and apply titles that visually relate to each artwork.
Nicknamed ‘The Mad One’ for her enthusiastic approach to graffiti art, Claudia Walde works under the name MadC as a graffiti artist and painter based in Germany. Internationally recognized for her vibrant and arresting public works on a monumental scale, MadC’s nearly 30-year career boasts murals across the world. Though never abandoning her street art practice, MadC has produced many of her new compositions on canvas, taking the influence of graffiti into the studio and away from its long-established place in the streets. Given MadC’s interest in challenging and exploring the potentiality of graffiti outside its original context and form, the digital space of NFTs is the next logical setting for her experimentation.
"The NFT collection can clearly be identified as my work but at the same time it is also clear that it isn’t something I have painted. That’s exactly what I wanted to achieve, a new layer to my work, a new little universe, something that includes chance and complexity and energy but from a new angle."
Color Rhythms is a digitally constructed and visually intoxicating collection, comprising deliberate compositions created through the spontaneous arrangement of color and form. A sophisticated generative algorithm was used to recreate MadC’s real-life interventions with her canvases, including the application of color, brushstrokes, paint spray, paint splashes and paint drips. One of the remarkable aspects of this drop is how closely these digital creations mirror the style of the physical works that MadC creates.
MadC worked closely with the HENI Studio over the course of many months in order to create the generative algorithm. MadC then hand-selected 1,000 images from the thousands that had been generated, to form her genesis NFT collection.
The generative algorithm was written in p5.js, a JavaScript library, and the code is stored on-chain. This means that all of the information required to recreate the images is stored within each NFT and on the blockchain itself. For ease of compatibility with other sites, the resulting artworks are also stored on IPFS, which is a decentralized storage protocol.
Each of the 1,000 artworks has a unique title which was created through the application of machine learning to a set of existing movie titles. Once the new titles had been generated, a separate machine learning model was used to match titles to images. This model was trained on millions of existing image and caption pairings. The model descriptively paired artworks with the titles that most aptly reflect their content.
MadC has an established practice of naming her works using movie characters and references to pop culture. In pairing these generative titles with the Color Rhythms NFTs, MadC evokes within each image fictional worlds, provocatively encouraging viewers to find narrative within abstraction.
The nuances of MadC’s painterly practice emerge in Color Rhythms through the variation in images. This allowed the hand of the artist to be captured and replicated digitally – from the instinctual choice of color and chance splashes of paint, to the sharp pull of a paintbrush on canvas. The variations between artwork images have been quantified and labeled within the following rarity categories: Number of Colors, Density, Forms, Monochrome and Color. Each of these categories has a number of different parameters which are used to make up the expressive forms and evocative color palettes that characterize this collection.