I have been following this paper for face anti spoofing and had executed patch based cnn but finding some difficulty in extracting features for depth based CNN. If you have solved that part can you please help me regarding that?
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 you: https://github.com/YadiraF/PRNet
I have been following this paper for face anti spoofing and had executed patch based cnn but finding some difficulty in extracting features for depth based CNN. If you have solved that part can you please help me regarding that?