Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/lemonyte/pyautotrace
Python bindings for AutoTrace, a bitmap to vector conversion library.
https://github.com/lemonyte/pyautotrace
autotrace bindings bitmap cython graphics image-processing pypi-package python svg tracing vector vector-graphics vectorization
Last synced: 18 days ago
JSON representation
Python bindings for AutoTrace, a bitmap to vector conversion library.
- Host: GitHub
- URL: https://github.com/lemonyte/pyautotrace
- Owner: lemonyte
- License: lgpl-2.1
- Created: 2022-09-26T04:38:32.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2025-01-02T03:19:43.000Z (about 2 months ago)
- Last Synced: 2025-01-02T04:23:20.512Z (about 2 months ago)
- Topics: autotrace, bindings, bitmap, cython, graphics, image-processing, pypi-package, python, svg, tracing, vector, vector-graphics, vectorization
- Language: Cython
- Homepage: https://pypi.org/project/pyautotrace
- Size: 181 KB
- Stars: 14
- Watchers: 2
- Forks: 4
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE.txt
Awesome Lists containing this project
README
# PyAutoTrace
Python bindings for [AutoTrace](https://github.com/autotrace/autotrace).
## Requirements
- [Python 3.9](https://www.python.org/downloads/) or higher
## Installation
Install PyAutoTrace using your package manager of choice.
```shell
python -m pip install pyautotrace
```
```shell
uv add pyautotrace
```
## Usage
```python
import numpy as np
from autotrace import Bitmap, VectorFormat
from PIL import Image
# Load an image.
image = np.asarray(Image.open("image.jpeg").convert("RGB"))
# Create a bitmap.
bitmap = Bitmap(image)
# Trace the bitmap.
vector = bitmap.trace()
# Save the vector as an SVG.
vector.save("image.svg")
# Get an SVG as a byte string.
svg = vector.encode(VectorFormat.SVG)
```
## Building
If you wish to build the package from source, the easiest way to do so is with [uv](https://docs.astral.sh/uv/).
Clone the repository and run the following commands inside the project directory.
```shell
# Clone the AutoTrace submodule.
git submodule update --init
# If you're on Windows, extract the GLib headers archive.
Expand-Archive "third-party\autotrace\distribute\win\3rdparty\glib-dev_2.34.3-1_win64.zip" -DestinationPath "third-party\glib"
# If you're on macOS, install GLib with Homebrew.
brew install glib
# Build the package with uv.
uv build
```
On Linux and macOS compilation requires GLib, pkg-config, and unzip to be installed on your system, which most Linux distributions include by default.
You can install GLib on macOS with `brew install glib`.
On Windows, in order to compile the generated C code, you will need to have
[Visual C++ Build Tools](https://visualstudio.microsoft.com/visual-cpp-build-tools/) or another C/C++ compiler installed.
## TODO
- Tests
- Documentation
## License
This project is licensed under the [LGPLv2.1](LICENSE.txt) license.
This project depends on the AutoTrace project, which is licensed under the [LGPLv2.1](https://github.com/autotrace/autotrace/blob/master/COPYING.LIB) license.
AutoTrace, and by extension this project, requires the presence of GLib to compile, which is licensed under the [LGPLv2.1](https://github.com/GNOME/glib/blob/main/COPYING) license, but this project does not depend on GLib to run.
This project contains code that replaces portions of [AutoTrace](https://github.com/autotrace/autotrace) and [GLib](https://github.com/GNOME/glib), defined in [`overrides.cpp`](autotrace/overrides.cpp). Some of the implementations were taken directly from, or are based on, the source code of their respective libraries.