https://github.com/efefurkankarakaya/reproducer

Image reproduction with genetic algorithm
https://github.com/efefurkankarakaya/reproducer

genetic-algorithm image-reproduction python3

Last synced: about 2 months ago
JSON representation

Image reproduction with genetic algorithm

• Host: GitHub
• URL: https://github.com/efefurkankarakaya/reproducer
• Owner: efefurkankarakaya
• Created: 2022-12-07T17:09:36.000Z (over 2 years ago)
• Default Branch: master
• Last Pushed: 2022-12-07T20:07:35.000Z (over 2 years ago)
• Last Synced: 2025-01-18T07:43:56.076Z (4 months ago)
• Topics: genetic-algorithm, image-reproduction, python3
• Language: Python
• Homepage:
• Size: 1.46 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Reproducer

Reproducer is an image reproduction project with genetic algorithm. It's extended and improved implementation (more accurate and faster reproduction for all colors) of ahmetdfgad/GARI project.

## Installation

- Clone the repository.

- `$ cd reproducer && virtualenv venv`

- `$ source venv/bin/activate`

- `$ pip install -r requirements.txt`

- Specify an image path in the code (which is in main function).

- Specify width and height multiplier in main function if you want to resize picture or have faster iterations.

- Specify the iteration count, the default is `10m`.

- `$ python3 reproducer.py`

- You can find outputs in `points/` folder.

## Strategies

- If you lower your picture resolution, you'll have more accurate reproduced images. Also `reproducer` will work faster.

- If you want to have faster reproduction and also same resolution, you can use upscalers like image-super-resolution.

- You can reproduce your friends' photos and let them use their reproduced photos as profile photo! That's funny, strongly recommended.

- Mostly, below 400x400 images are over 90% percentage being reproduced well in `5m` iterations.

## Example (Thanks to Elon Musk)

  Iterations: 0

  Iterations: 5k

  Iterations: 10k

  Iterations: 50k

  

  

  

  

  Iterations: 100k

  Iterations: 250k

  Iterations: 500k

  Iterations: 1m