https://github.com/iver56/emoji-art-generator

Use a genetic algorithm to evolve an image by putting emojies on a canvas
https://github.com/iver56/emoji-art-generator

art emoji genetic-algorithm mosaic mosaic-images

Last synced: 3 months ago
JSON representation

Use a genetic algorithm to evolve an image by putting emojies on a canvas

• Host: GitHub
• URL: https://github.com/iver56/emoji-art-generator
• Owner: iver56
• Created: 2019-02-05T20:40:28.000Z (over 6 years ago)
• Default Branch: master
• Last Pushed: 2022-08-20T05:47:39.000Z (almost 3 years ago)
• Last Synced: 2025-02-28T23:12:21.142Z (4 months ago)
• Topics: art, emoji, genetic-algorithm, mosaic, mosaic-images
• Language: Python
• Homepage:
• Size: 14.2 MB
• Stars: 78
• Watchers: 3
• Forks: 8
• Open Issues: 1
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Emoji Art Generator

| Description | Image |

| ----------- | ----- |

| Target image | ![Target image](data/target_images/sunglasses2.png) |

| RGB MSE fitness | ![Evolved emoji image](demo/rgb_mse.gif) |

| LAB MSE fitness | ![Evolved emoji image](demo/lab_mse.gif) |

| LAB Delta E CIE 2000 fitness | ![Evolved emoji image](demo/lab_delta_e.gif) |

| LAB Delta E CIE 2000 combined with SSIM fitness | ![Evolved emoji image](demo/lab_delta_e_ssim.gif) |

| SSIM fitness | ![Evolved emoji image](demo/ssim.gif) |

# Setup

* `conda env create`

# Usage

Prepare a target image (store it in `data/target_images/`) and a set of emojies/tiles (store them in `data/emoji/`). Then run the following command:

`python -m app.generator.generate`

```

usage: generate.py [-h] [--starting-canvas STARTING_CANVAS] [--target TARGET]

                   [--fitness {RGBMSE,LABDeltaESSIM,LABMSE,LABDeltaE,SSIM}]

                   [-g NUM_GENERATIONS] [-p POPULATION_SIZE] [--width WIDTH]

                   [--height HEIGHT] [--emoji-size EMOJI_SIZE]

optional arguments:

  -h, --help            show this help message and exit

  --starting-canvas STARTING_CANVAS

                        Path to an image to start with. If not specified, a

                        white image will be used.

  --target TARGET       Filename of target image. Should reside in

                        data/target_images/

  --fitness {RGBMSE,LABDeltaESSIM,LABMSE,LABDeltaE,SSIM}

                        Choose fitness evaluator. See fitness.py for more

                        information.

  -g NUM_GENERATIONS, --num-generations NUM_GENERATIONS

  -p POPULATION_SIZE, --population-size POPULATION_SIZE

  --width WIDTH         If specified, resize the target image (and the

                        starting canvas) to this width. Otherwise, keep the

                        original width.

  --height HEIGHT       If specified, resize the target image (and the

                        starting canvas) to this height. Otherwise, keep the

                        original height.

  --emoji-size EMOJI_SIZE

```