README.md

Models

We provide several publicly available lightweight models for face recognition and facial expression recognition that obtained high accuracy of LFW (Labeled Faces in-the-Wild) and AffectNet datasets.

Mobile demo

The demo Android application is available at our Google drive. Here is the result of its running on my personal device:

[]

Research papers

If you use our models, please cite the following papers:

@incollection{savchenko2024device,
  title={Device-Specific Facial Descriptors: Winning a Lottery with a SuperNet},
  author={Savchenko, Andrey and Maslov, Dmitry and Makarov, Ilya},
  booktitle={ECAI},
  pages={4439--4442},
  year={2024},
  publisher={IOS Press}
}

@article{savchenko2024autoface,
  author={Savchenko, Andrey V.},
  journal={IEEE Access}, 
  title={{AutoFace}: How to Obtain Mobile Neural Network-Based Facial Feature Extractor in Less Than 10 Minutes?}, 
  year={2024},
  volume={12},
  pages={25106-25118},
  doi={10.1109/ACCESS.2024.3365928}}

@article{savchenko2023fast,
  author={Savchenko, Andrey V. and Savchenko, Lyudmila V. and Makarov, Ilya},
  journal={IEEE Access}, 
  title={Fast Search of Face Recognition Model for a Mobile Device Based on Neural Architecture Comparator}, 
  year={2023},
  volume={11},
  pages={65977-65990},
  doi={10.1109/ACCESS.2023.3290902}
}