Metadata-Version: 2.3
Name: sketchy-svg
Version: 0.1.0
Summary: A png drawing to svg converter
Requires-Dist: anywidget>=0.9.21
Requires-Dist: marimo>=0.20.1
Requires-Dist: matplotlib>=3.8.0
Requires-Dist: mpld3>=0.5.12
Requires-Dist: networkx>=3.4.2
Requires-Dist: numpy
Requires-Dist: pillow>=10.2.0
Requires-Dist: pytest>=9.0.2
Requires-Dist: python-dotenv>=1.2.1
Requires-Dist: scikit-image>=0.20.0
Requires-Dist: traitlets>=5.14.3
Requires-Dist: onnxruntime>=1.22.1 ; extra == 'onnx'
Requires-Python: >=3.10
Provides-Extra: onnx
Description-Content-Type: text/markdown

# sketch-vectorization

![](./images/illustration_pipeline.png)

## Goal

The goal of this project is to implement the following paper: <https://www-sop.inria.fr/reves/Basilic/2016/FLB16/fidelity_simplicity.pdf>

We also implemented several new ideas, such as a [convolutional neural network](./notebooks/cnn.ipynb) with synthetic data augmentation for preprocessing.

## Use

To install the dependencies, install [uv](https://github.com/astral-sh/uv) and run `uv sync`.

Then, you can run `uv run src/bez/app.py some_image.png`. This will create the png in the current directory.

## Documentation

You can take a look at:
- the [Walkthrough](./documentation/walkthrough.ipynb)
- the [Presentation](./documentaion/report.pdf)
