Overview

📐 Convert images to Low-Poly art using Delaunay triangulation.

Table of contents

1. Installation
2. Usage
3. Use as a library
4. Sample
5. License

Installation

You need Python 3.10 or higher.

I strongly recommend to use virtual environment such as Anaconda. You can download Anaconda here.

Follow manual below to create conda virtual environment for Triangler with the Anaconda.

$ conda create -n triangler python=3.12
$ activate triangler
(triangler)$ python -m pip install git+https://github.com/tdh8316/triangler/

Usage

(triangler)$ python -m triangler -h
positional arguments:
  input                 Input image

options:
  -h, --help            show this help message and exit
  -o OUTPUT, --output OUTPUT
                        Output image name
  -p POINTS, --points POINTS
                        Number of sample points to use
  -e {canny,entropy,sobel}, --edge-detector {canny,entropy,sobel}
                        Edge detection algorithm
  -s {poisson_disk,threshold}, --sampler {poisson_disk,threshold}
                        Point sampling algorithm
  -r {centroid,mean}, --renderer {centroid,mean}
                        Color polygon rendering algorithm
  -l, --reduce          Reduce the result image size to match the input image
  -V, --version         show program's version number and exit
  -d, --debug           Enable debug mode

API

You can also use Triangler as a library.

import triangler

triangler.convert(
    img="INPUT_IMAGE.jpg",
    save_path="OUTPUT_IMAGE.jpg",
)

Sample

Effect of the number of points

Original Image	5000 Points
1000 Points	500 Points

More samples

Original	Triangler

License

Licensed under the MIT License.

Copyright 2024 Donghyeok Tak

Credits

Some algorithms, including the Poisson disk sampling, are based on pmaldonado/PyTri.