deepcam-cn / Face-Anti-spoofing.pytorch

Train code of face anti-spoofing with a single RGB frame

Home Page:https://deepcam.com/

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Training on new dataset

bcba25 opened this issue · comments

Currently I'm trying to train the network using id pro dataset, but it seems like inference is not really imrpoving, i basically got this kind of result:

threshold 0.5: frp = 0.07120253164556962 ; tpr = 0.08847184986595175

Could you suggest on how can I improve it ? Before the training I crop the faces, scaled it to 256x256, labeled it correctly, used your dataset as an example. Your is training pretty well