Oeslle Lucena

Hi @apollo-time! You're welcome! So, what do you mean about train image dataset? Coul you clarify it? I used REPLAY-ATTACK and 3DMAD databases for training and testing.

Hi @zhouhui1992 ! Unfortunately I cannot send you a copy of the data since I signed and EULA form to dot not distribute them. You can get the data on...

Hi @aakashns , first of all, thank you for ur interesting in our work. I used the OpenFace algorithm, which is open source. I did not change anything in the...

@okmmsky888 Yep, they are already trained and they can be used directly. No need to combine with VGG-16 weights. The two documents is because they were fine-tuned for both datasets...

@okmmsky888 It's a mismatch of shapes in the network. ValueError: ('shapes (1,25088) and (4608,256) not aligned: 25088 (dim 1) != 4608 (dim 0)', (1, 25088), (4608, 256)) Are you loading...

@okmmsky888 Now I see the problem. I forgot to mention on github but it is explained on the paper that the trained models are for size 96x96, since I did...

@okmmsky888 I set 0 = real, 1= attack in both models, REPLAY and 3DMAD. 3DMAD is for database contained impostors using fake masks and REPLAY-ATTACK has more types of attacks....

@okmmsky888 what do you mean by "does not seem to be ideal"? is the CNN giving a wrong classification? Also, from which database this image came from? Remember that tests...

@okmmsky888 I mentioned before that test with different databases are not easy and what you're doing is not correct. Therefore, that result is expected .

You have to do a training step using different databases to properly classify images from a different dataset and validate your system. I would recommend you to search about inter...