Exploring Bridges Between Algorithmic and AI-generated Art

TL;DR

This work addresses bridging algorithmic art and AI-generated art within a creative coding environment by introducing two libraries: GenP5, which enables diffusion-based stylization and conditioning of algorithmic art in p5.js, and P52Style, which learns the style of algorithmic outputs to guide AI generation. GenP5 provides per-buffer diffusion stylization and conditioning via a client-server pipeline, enabling real-time stylization and pattern-driven constraints. P52Style adds a GUI-driven workflow to capture algorithmic-art style, learn it through Visual Style Prompting, and apply the learned style to AI-generated outputs, addressing the challenge of style transfer from algorithmic artifacts. Together, these tools demonstrate viable integration of programmable art with generative AI, enabling new forms of creative expression and setting the stage for broader, real-time, cross-domain workflows in creative coding environments.

Abstract

In this paper, we bridge algorithmic and AI art by adding functionality to the creative coding environment. We create two systems that demonstrate how AI features can enhance algorithmic art and, conversely, how AI art can be styled based on algorithmically-generated artifacts. The first library, GenP5, extends p5.js to allow the artist to apply diffusion models to style and 'condition' their algorithmically-constructed art. The second, P52Style, can learn the 'style' of an algorithmically generated artifact and apply that when creating new AI art.

Figures (6)

• Figure 1: Left: GenP5 method overview. Numbers indicate a frame index. Right: UI elements that will be dynamically created when GenP5 library is used. In this example, a slight background pattern is drawn in the background canvas. Two stylized buffers are created, each containing a ring-shape animation. One nonstylized buffer contains bubble effect filters (this buffer is not displayed in the UI).
• Figure 2: Code snippet of creating art projects with GenP5 library.
• Figure 3: Three simple examples of generative art project using GenP5 functions for stylization. Prompts for stylize buffers: 'lightblue human neurons of radial shapes' (row I), 'abstract galaxy pattern' (row II left), 'an abstract total solar eclipse' (row II right), 'realistic flower stems' (row III left), 'realistic purple roses' (row III right).
• Figure 4: Four examples of generative art project using GenP5 functions for conditioning. The input contour maps are inverted.
• Figure 5: Overview of p52style structure.