Metadata-Version: 2.1
Name: facelandmarks
Version: 0.1.0
Summary: Human face landmarking via machine learning model
Author: JaroslavTrnka
Author-email: <jaroslav_trnka@centrum.cz>
License: MIT
Keywords: python,pytorch,face,landmarks,machine learning
Classifier: License :: OSI Approved :: MIT License
Classifier: Development Status :: 4 - Beta
Classifier: Programming Language :: Python :: 3.10
Classifier: Operating System :: OS Independent
Description-Content-Type: text/markdown
License-File: LICENCE
Requires-Dist: glob2 ==0.7
Requires-Dist: imagesize ==1.4.1
Requires-Dist: ipykernel ==6.25.1
Requires-Dist: ipython ==8.14.0
Requires-Dist: jupyter-client ==8.3.0
Requires-Dist: jupyter-core ==5.3.1
Requires-Dist: matplotlib ==3.7.2
Requires-Dist: matplotlib-inline ==0.1.6
Requires-Dist: mediapipe ==0.10.3
Requires-Dist: nbformat ==5.9.2
Requires-Dist: numpy ==1.25.2
Requires-Dist: opencv-contrib-python ==4.8.0.74
Requires-Dist: opencv-python ==4.8.0.74
Requires-Dist: packaging ==23.1
Requires-Dist: pandas ==2.1.1
Requires-Dist: Pillow ==10.0.0
Requires-Dist: plotly ==5.17.0
Requires-Dist: requests ==2.31.0
Requires-Dist: scikit-learn ==1.3.1
Requires-Dist: scipy ==1.11.3
Requires-Dist: torch ==2.1.1
Requires-Dist: torchmetrics ==1.2.0
Requires-Dist: torchvision ==0.16.1
Requires-Dist: tqdm ==4.65.0
Requires-Dist: urllib3 ==2.1.0

# Face-Landmarking
OpenCV and Pytorch based system for finding face landmarks coordinates. Based on LBF and MediaPipe models.
LBF and MediaPipe models are linearly projected and CNN finetuned into the system of landmarks used in:
Kleisner, K., TureÄŤek, P., Roberts, S.C. et al. How and why patterns of sexual dimorphism in human faces vary across the world. Sci Rep 11, 5978 (2021). https://doi.org/10.1038/s41598-021-85402-3

![image](https://github.com/JaroslavTrnka1/Face-Landmarking/assets/85038909/864c4de6-0dbf-46a5-aa3e-9c821bfb41a8)



!!!
![image](https://github.com/JaroslavTrnka1/Face-Landmarking/assets/85038909/df2b2c53-3d32-46f9-af3c-3561b1e64bd8)
