Update (March, 2023): The whole dataset-SuHiFiMask is released now. If you are interested in this dataset, you will be required to first sign a licensing agreement (link) and send it to: Jun Wan, jun.wan@ia.ac.cn  . 

How to participate this challenge and dataset download?

All participants (using formal email, such as company/university email, not Gmail, sina.com, qq.com...) should provide their TeamName in Codalab, and company/university name first to Jun Wan (jun.wan@ia.ac.cn). The maximum number of teams from the same institute is three, and the team members should be mutual which means one participant can only join in one team.  After the application is approved, we will share the available download link for this contest only. If you need this dataset for academic research future, you have to apply again after this contest. 

Note: Only the approved TeamName can take part in this contest. And the final ranking results will release when this contest is over. Even some TeamNames have not submitted their code/results, we will still release their scores (maybe the score is zero).

The email should contain the following information:

=================================

subjects: Dataset application for the 4th Face Anti-spoofing Challenge@CVPR2023

Team Name: XX 

Team Leader: XX 

Team Members: XX,XX,XX (maximum 5 persons per team)

Homepage:    (the formal website of the homepage of your workplace (i.e., university/company/institute)).

=================================

If your application is approved, you will be required to first sign a licensing agreement and then we will share the available download link.

Q&A

Q: What does 0 and 1 of the label mean?

A: live: 1 fake:0