fb-wh / face-anti-spoofing

ChaLearn Face Anti-spoofing Attack Detection Challenge@CVPR2019

Face-anti-spoofing

ChaLearn Face Anti-spoofing Attack Detection Challenge@CVPR2019

Usage: 运行环境： Ubuntu 16.04, GPU:GTX 1080Ti

需要库： python == 3.6.2 pytorch == 0.4 .1 numpy ==1.15.2 matplotlib == 3.0.2 sklearn ==0.20.3 opencv-python==3.4.3 Pillow

安装相关库文件：

pip install -r requirements.txt

数据集文件夹如下所示：

移除了预训练模型！

步骤：

1. cd Face-anti-spoofing
2. pip install -r requirement.txt #安装所需库
3. 运行: CUDA_VISIBLE_DEVICES=0 python train.py --root datapath

运行： CUDA_VISIBLE_DEVICES= 0 python train_xx.py --root /home/xxxx/datasets/CASIA-SURF