Image to Pixels
What survives when you reduce an image?
A browser workbench for pixels, dithering and ASCII. Compare treatments, edit palettes and save reproducible recipes.
The problem
Reducing an image to a grid and a few colors changes its shape, brightness and texture. A useful experiment lets you compare those choices and reproduce the result.
The approach
- Use a shared image pipeline for pixels, Bayer and error-diffusion dithering, and fixed bitmap ASCII glyphs.
- Freeze treatment A while editing B. Inspect actual pixels, adjust or extract a palette, and export a PNG.
- Record every setting, the palette, the grain seed and the source identity in a versioned recipe. Process images locally in a browser worker.
What you can inspect
Open one of three procedural examples or your own image. The repository includes deterministic renderer tests, recipe round trips, alpha handling and regression tests for the original Python converter.
Scope & limitations
Still images only, processed on your device. Sources are normalized to 1600 pixels on the longest edge; recipes reference user images without including them. Regional painting and layers are outside this first release.
Run it locally
Node.js 22 or newer. The default lab needs no API key or cloud account.
git clone https://github.com/alejandrofrank/art_exploring.git
cd art_exploring
npm run devThen open http://127.0.0.1:4314. Follow the repository for optional integrations and verification commands.