search

ZhangYuanhan-AI / CelebA-Spoof

[ECCV2020] A Large-Scale Face Anti-Spoofing Dataset
531 stars 92 forks source link

A question about the crop area for AENET #40

Closed kadirbeytorun closed 3 years ago

kadirbeytorun commented 3 years ago

Hello david,

I wanted to ask about what should be the crop area be like as aenet input? Should be the direct output of face detector? Or are we supposed to crop to include whole hair, neck area etc? Because I observed there are face attributes in celeba such as "wearing necklace, wearing necktie, gray hair" Which cannot be seen if I crop only the face area.

ZhangYuanhan-AI commented 3 years ago

From our detailed experiments, the performance becomes worse as the cropped area become larger. Therefore, we release our model trained based on the cropped face area finally.

kadirbeytorun commented 3 years ago

Thanks for your rapid response mate. So since only cropped face will be the input of the network, thus face attribute results such as wearing tie, hat etc. are not reliable I guess.