This script converts an (ideally) grayscale image into a autostereogram of the same dimensions as the input image. White in the input image corresponds to closer to the viewer, while black or transparent corresponds to farther.
- Ensure Numpy and PILLOW are installed.
- Run
python3 main.py <pathToInputImage. - The output image will created in the same directory as the input image and with the name "generatedNoise.png"
- Despite multiple revisions to the algorithm, I still don't know if I'm generating the image properly, so the result might not look that great.
- The column width is hardcoded to 50 pixels (with a minimum of 5 columns).
- The noise is composed of random colors instead of taken from a palette with a generally coherent hue like most other images of this type.
- There is no way to produce a result of different resolution than the input image.
Don't. I just made this project for fun and don't really feel like handling any PRs. I mainly made this repo public in hopes that someone finds it neat.