NTU-ESOS-Capstone topics of Engineering Science and Ocean Engineering
This is an implementation of AI-based Face Anti-Spoofing(FAS) Detection. The goal of this project is to perform FAS detection on Rasberry Pi 4. We trained CDCN++ which was proposed in "Searching Central Difference Convolutional Networks for Face Anti-Spoofing" from scratch. The model was trained on OULU-NPU and SiW dataset.
depth.map.mp4
(kind of laggy due to the limited computation resources on Rasberry Pi)
Raspberry.pi4-FAS.mp4
-FAS.mp4
-
Weights
Weights can be acquired here.
-
Start detecting!
$ python run.py --detector_path --model_weights
Use
--detector_path
and--model_weights
to assign paths.
@inproceedings{yu2020searching,
title={Searching Central Difference Convolutional Networks for Face Anti-Spoofing},
author={Yu, Zitong and Zhao, Chenxu and Wang, Zezheng and Qin, Yunxiao and Su, Zhuo and Li, Xiaobai and Zhou, Feng and Zhao, Guoying},
booktitle= {CVPR},
year = {2020}
}
@inProceedings{feng2018prn,
title = {Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Network},
author = {Yao Feng and Fan Wu and Xiaohu Shao and Yanfeng Wang and Xi Zhou},
booktitle = {ECCV},
year = {2018}
}
31