yaojieliu / ECCV20-STDN

Source code for ECCV 2020 paper: On Disentangling Spoof Trace for Generic Face Anti-Spoofing

Home Page:http://cvlab.cse.msu.edu/project-face-anti.html

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Need suggestion for input

GayatriPurandharT opened this issue · comments

commented

The input is both cropped face image and it's npy representation. Is there any specific program you used to generate crop and corresponding .npy file?
I did check your github link- https://github.com/yaojieliu/ICCVW2017-DenseFaceAlignment
I can't access the Basel Face Model (BFM) page that has been shared in the github page.
Also, does the 'DenseFaceAlignment' generate the corresponding npy for each input?
Thank you very much.

You can also try https://github.com/1adrianb/face-alignment to obtain the landmarks.
.npy file is basically a file to save numpy array. You can use numpy.save() and numpy.load() to access the .npy file.

commented

Thank you for the clarification.
May I know the algorithm you used for cropping the faces from SiW-M video dataset?