Face anti-spoofing is essential to prevent face recognition systems from a security breach. Much of the progresses have been made by the availability of face Anti-spoofing benchmark datasets in recent years. However, existing face Anti-spoofing benchmarks have limited number of subjects(≤170) and modalities (≤2), which hinder the further development of the academic community.
To facilitate future face Anti-spoofing research, we release a large-scale multi-modal dataset, namely Chalearn CASIA-SURF, which is the largest publicly available dataset for face Anti-spoofing in terms of both subjects and visual modalities. Specifically, it consists of 1, 000 subjects with 21, 000 videos and each sample has 3 modalities (i.e., RGB, Depth and IR).
More information can be found in our paper:
- Shifeng Zhang, Xiaobo Wang, Ajian Liu, Chenxu Zhao, Jun Wan, Sergio Escalera, Hailin Shi, Zezheng Wang, Stan Z. Li, " CASIA-SURF: A Dataset and Benchmark for Large-scale Multi-modal Face Anti-spoofing ", arXiv, 2018.
If you are interested in the CASIA-SURF dataset or the challenge in CVPR workshop 2019, please cite this paper.
How to participate
- Register for the Chalearn Face Spoofing Attack Detection Challenge (codaLab website).
- Download, read the Contest Rules, and sign the agreement (link).
- Send the your signed agreements to: Jun Wan, email@example.com
Participants should submit a ZIP file containing only one text file named exactly as required below (do not add any folder in the ZIP). The ZIP file can be named randomly. Each line in the txt file contains four parts and each part is separated by a space. The first three parts are the relative paths of the three modes of each test sample, and the last part is the prediction score that the sample belongs to the real sample.
For Example(Evaluation requirements scores are retained at least 6 decimal places):
Testing/0000/000006-color.jpg Testing/0000/000006-depth.jpg Testing/0000/000006-ir.jpg 0.9851337671279907
Testing/0000/000007-color.jpg Testing/0000/000007-depth.jpg Testing/0000/000007-ir.jpg 3.7812408208992565e-06
Testing/0000/000008-color.jpg Testing/0000/000008-depth.jpg Testing/0000/000008-ir.jpg 6.042769018677063e-05
Testing/0000/000009-color.jpg Testing/0000/000009-depth.jpg Testing/0000/000009-ir.jpg 0.0018127199728041887
Testing/0000/000010-color.jpg Testing/0000/000010-depth.jpg Testing/0000/000010-ir.jpg 1.0
Testing/0000/000011-color.jpg Testing/0000/000011-depth.jpg Testing/0000/000011-ir.jpg 1.262649857380893e-05