Face Spinner is an application for mitigating face orientation problems in image processing.
# For the current version, please execute this first
export PIP_FIND_LINKS=https://download.pytorch.org/whl/torch_stable.html
# If pipenv is installed
pipenv sync
# If virtualenv or pip is preferred
pip install -r requirements.txt
python -m example.basic /path/to/your/image.jpg # Result saved to output.jpg
python application.py # Run on localhost:5000
# Check usage using `curl localhost:5000`
# Test making a request to align an image. Result will be outputted to output.jpg
curl -X POST -F "image=@tests/img/face-270.jpg" --output output.jpg localhost:5000/align
docker build -t face-spinner .
docker run -p 5000:5000 face-spinner # Run on localhost:5000
# Check usage using `curl localhost:5000`
# Test making a request to align an image. Result will be outputted to output.jpg
curl -X POST -F "image=@tests/img/face-270.jpg" --output output.jpg localhost:5000/align
- Pytorch implementation of PCN by siriusdemon
- Real-Time Rotation-Invariant Face Detection with Progressive Calibration Networks
@inproceedings{shiCVPR18pcn,
Author = {Xuepeng Shi and Shiguang Shan and Meina Kan and Shuzhe Wu and Xilin Chen},
Title = {Real-Time Rotation-Invariant Face Detection with Progressive Calibration Networks},
Booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
Year = {2018}
}