Ecosyste.ms: Awesome

An open API service indexing awesome lists of open source software.

Awesome Lists | Featured Topics | Projects

https://github.com/Jzou44/photo2pixel

Algorithm used in photo2pixel.co to convert photo to pixel style(8-bit) art.
https://github.com/Jzou44/photo2pixel

image-processing pixel pixel-art pixelart pytorch

Last synced: 3 days ago
JSON representation

Algorithm used in photo2pixel.co to convert photo to pixel style(8-bit) art.

Awesome Lists containing this project

README

        

# ![LOGO](images/doc/favicon-original.png) Photo2Pixel

---
English | [简体中文](./README_cn.md)

[Online Tool](https://photo2pixel.co) |
[Colab](https://colab.research.google.com/drive/108np4teybhBXHKbPMZZ1fykDuUeF2aw8?usp=sharing) |
[Tutorial](#Tutorial)

photo2pixel is an algorithm converting photo into pixel art. There is an [online converter photo2pixel.co](https://photo2pixel.co)
. you can try different combination of pixel size and edge threshold to get the best result.

mountain 8bit style pixel art
holy temple 8bit style pixel art

## Prerequisites
- python3
- pytorch (for algorithm implementation)
- pillow (for image file io)

## Tutorial
---
photo2pixel is implemented with Pytorch, the easiest way to run it is [Colab](https://colab.research.google.com/drive/108np4teybhBXHKbPMZZ1fykDuUeF2aw8?usp=sharing),
or you can run it with command as bellow:
```bash
# use default param
python convert.py --input ./images/example_input_mountain.jpg

# or use custom param
python convert.py --kernel_size 12 --pixel_size 12 --edge_thresh 128
```

| Parameter | Description | Range | Default |
|-------------|:-------------------------------------------------------------------------:|:-----------:|:-----------------------------------:|
| input | input image path | / | ./images/example_input_mountain.jpg |
| output | output image path | / | ./result.png |
| kernel_size | larger kernel size means smooth color transition | unlimited | 10 |
| pixel_size | individual pixel size | unlimited | 16 |
| edge_thresh | the black line in edge region, lower edge threshold means more black line | 0~255 | 100 |