shicaiwei123
shicaiwei123
训练数据10w,1k图片测试准确率96% 把作者CNN特征提取的dropout去掉,只留下Linear层的dropout, 多增加几层layer,但是保持图片尺寸衰减还是8倍,也就是,只有3个maxpooling层, 余弦学习率衰减,
Sorry for the late reply, I missed the message before. Could you describe your difficulty in more detail. If it is the depth map estimation, maybe the PRNet can help...
sorry for that，it has been updated now
Hello, sorry for the late reply. The model depth_patch.pth has been uploaded just now. It's worth noting that, here, only the depth map estimation is implemented and the SVM has...
The performance post on the github only leverage the patch. But we also provide the code for depth model At 2022-05-20 14:36:36, "Shobhit Sharma" ***@***.***> wrote: @shicaiwei123 is it only...
> Hi, thanks for sharing your code, could you update README on how to run, train and test your method. Thanks Hi，I am sorry to reply so late.i have add...
> @shicaiwei123 I see. Thanks for reply, are you still plannign to implement any anti spoofing solutions ? If yes what kind are you planning to work with ? uh-huh,yeah,actually,...
是的，living face 对应是1，我会更正一下。谢谢。 电子科技大学 信息与通信工程院 School of Information and Communication Engineering University of Electronic Science and Technology of China (UESTC) +18285516726 韦仕才 发自 网易邮箱大师 ---- Replied mail ---- | From...
Sorry for replying so late, we haven't tried it on unix, and it seems that this problem will also occur on Windows.
这个不影响使用,等一会,就会出现程序运行结果