FeatherNets_Face-Anti-spoofing-Attack-Detection-Challenge-CVPR2019 icon indicating copy to clipboard operation
FeatherNets_Face-Anti-spoofing-Attack-Detection-Challenge-CVPR2019 copied to clipboard

Code for 3rd Place Solution in Face Anti-spoofing Attack Detection Challenge @ CVPR2019,model only 0.35M!!! 1.88ms(CPU)

Results 81 FeatherNets_Face-Anti-spoofing-Attack-Detection-Challenge-CVPR2019 issues
Sort by recently updated
recently updated
newest added

Val——Do I need to download this file?And where to download it? ![image](https://user-images.githubusercontent.com/76394313/127651612-6ee9f9ee-49f6-479f-97bc-d77ff6942d50.png)

How to solve it? ![image](https://user-images.githubusercontent.com/76394313/127492104-4495c92c-4d31-4266-a815-5d8bfff7176f.png)

it seems the MobileNetV2 Model is invalid, could you upload again?

this is my output when i convert to openvino. I don't understand. can you help me ? ![image](https://user-images.githubusercontent.com/55668910/119986570-32c7ab00-bfee-11eb-8bdb-bb06195bbe76.png)

Does it apply to person re-identification?

你好,我对训练过程有两点疑问。 1·比赛公布数据集有RGB,IR和Depth数据,看你的代码是仅仅只用Depth数据训练吗?为什么选择不使用三种类型的数据一起训练分类器呢,混合训练是效果差吗? 2·样本增强方面。在data/our_filelist/2depth_train.txt中,sample-eye-nose_out是对原始深度图直接提取保留眼睛和鼻子部分然后其他区域都置为0吗?如果是这样处理,请问对于网络学习泛化能力提高明显吗? 谢谢分享!

请问视频数据集可以下载吗?

作者您好,请问你们是如何将原始的深度图像数据转换成0~255灰度图像数据的?