Metadata-Version: 2.1
Name: deface
Version: 0.1.0
Summary: Video anonymization by face detection
Home-page: https://gitlab.com/mdraw/deface
Author: Martin Drawitsch
Author-email: martin.drawitsch@gmail.com
License: UNKNOWN
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: imageio
Requires-Dist: imageio-ffmpeg
Requires-Dist: numpy
Requires-Dist: tqdm
Requires-Dist: scikit-image
Requires-Dist: opencv-python
Provides-Extra: gpu
Requires-Dist: onnxruntime-gpu ; extra == 'gpu'

Deface: Video anonymization by face detection
=====================================

Installation
------------

Current git master:

    $ pip install 'git+https://gitlab.com/mdraw/deface'

Usage
-----

After installing the `deface` package:

    $ deface -i <input.mp4> -o <output.mp4>

Without installation:

    $ python3 -m deface.deface -i <input.mp4> -o <output.mp4>


GPU acceleration
----------------

If you have a CUDA-capable GPU, you can enable automatic GPU acceleration by installing ONNX Runtime:

    $ pip install onnx onnxruntime-gpu

If it can be loaded sucessfully, it is automatically used for face detection.
s

Credits
-------

- centerface.{onnx,py}: based on https://github.com/Star-Clouds/centerface (8c39a497),
  MIT license (https://github.com/Star-Clouds/CenterFace/blob/36afed/LICENSE)

