PresSafe TF Implementation

PresSafe: Barometer-Based On-Screen Pressure-Assisted Implicit Authentication for Smartphones

unlike many works that use PyTorch as their backend, this project is implemented using TensorFlow 2.x.
the contributor of the code: Muyan Yao

---

Our implementation of PresSafe: Barometer-Based On-Screen Pressure-Assisted Implicit Authentication for Smartphones.

• This repository includes the implementation of the deep learning backbone network involved to extract features from the pre-processed user data.
• Should you have any concerns, feel free contact with me directly at muyanyao \at ieee.org

If you use PresSafe in your project or research, please cite the following paper:

• Muyan Yao, et al. (2023)

> M. Yao, D. Tao, R. Gao, J. Wang, S. Helal and S. Mao, <br/>
> "PresSafe: Barometer-Based On-Screen Pressure-Assisted Implicit Authentication for Smartphones," <br/>
> in IEEE Internet of Things Journal, vol. 10, no. 1, pp. 285-302, 1 Jan.1, 2023, doi: 10.1109/JIOT.2022.3199657. <br/>

```bibtex
@article{9858865,
  author={Yao, Muyan and Tao, Dan and Gao, Ruipeng and Wang, Jiangtao and Helal, Sumi and Mao, Shiwen},
  journal={IEEE Internet of Things Journal}, 
  title={PresSafe: Barometer-Based On-Screen Pressure-Assisted Implicit Authentication for Smartphones}, 
  year={2023},
  volume={10},
  number={1},
  pages={285-302},
  doi={10.1109/JIOT.2022.3199657}}

How to Install

The following dependency is required to have this project working normally:

• conda (anaconda, miniconda, or other variants)
• CUDA (if GPU based acceleration is preferred)

The python environment required for this project can be easily installed through:

conda env create -f pressafe.yaml

Have a nice day!