Face-Spoofing-Detection-or-Face-Liveness-Detection icon indicating copy to clipboard operation
Face-Spoofing-Detection-or-Face-Liveness-Detection copied to clipboard

How to train model

Open AblAzhidin opened this issue 7 years ago • 5 comments

Hello! Is there already trained SVMs we can use? And if I am going to train my own I need to specify the pass to txt file only, right? Like here: const string train_image_list = src_folder + "/train_image_list.txt". Also what should be the format of the txt file? Should it contain fake and real images together? Thank you!

AblAzhidin avatar Feb 08 '18 04:02 AblAzhidin

train_image_list is a txt file name that contains the image (both fake and real) name or relative path. if you want to train your own data, please run the trainSVM.cpp

yongw5 avatar Feb 28 '18 08:02 yongw5

Hi@wuyongchn , where is the image list, it seems there is no such files or dirs in this repository

anguoyang avatar Mar 02 '18 05:03 anguoyang

hello 请问你用的opencv是哪个版本的呀

quliulangle avatar May 09 '18 01:05 quliulangle

i think you should provide the whole files?

daixiangzi avatar Sep 08 '18 09:09 daixiangzi

Whether you provide models or not, we have to train ourselves, thank you for all your code. If trained model, it would be better.

sunjunlishi avatar Nov 14 '18 09:11 sunjunlishi