# edges2calligraphy

**Repository Path**: mirrors/edges2calligraphy

## Basic Information

- **Project Name**: edges2calligraphy
- **Description**: edges2calligraphy 使用名为 pix2pix 的神经网络将手写文字转换为中国书法字体（神经网络使用了大约 200 个赵孟頫的真迹草书文字进行训练）
- **Primary Language**: JavaScript
- **License**: MIT
- **Default Branch**: master
- **Homepage**: https://www.oschina.net/p/edges2calligraphy
- **GVP Project**: No

## Statistics

- **Stars**: 9
- **Forks**: 0
- **Created**: 2021-09-15
- **Last Updated**: 2026-01-24

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

![](screenshots/icon.png)

# edges2calligraphy
*From scribbles to Chinese calligraphy* （潦草字->赵体草书）
### Interactive demo: https://lingdong-.github.io/edges2calligraphy/

This project uses a neural network called [pix2pix](https://arxiv.org/pdf/1611.07004.pdf) to transfer your scribbles into Chinese Calligraphy. It is trained on ~200 cursive characters from [Zhao Mengfu](https://en.wikipedia.org/wiki/Zhao_Mengfu)'s *Thousand Character Classic in Regular and Cursive Script* (《赵孟頫真草千字文》), labeled using custom software. It is largely based on [affinelayer's edges2cats demo](https://affinelayer.com/pixsrv/) and [pix2pix-tensorflow](https://github.com/affinelayer/pix2pix-tensorflow) project, and uses [deeplearn.js](https://deeplearnjs.org) to perform GPU accelerated computation in your browser.

Try the online deomo: https://lingdong-.github.io/edges2calligraphy/. (Tested in Chrome and Firefox. Does not work in Safari.)

## Examples

![](screenshots/screen01.png)

## Training

This project uses [pix2pix-tensorflow](https://github.com/affinelayer/pix2pix-tensorflow) for training. See [`/tools/README.md`](/tools/README.md) for details.