CASIA-SURF: A Large-Scale Multi-Modal Benchmark for Face Anti-Spoofing

Abstract

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 antispoofing benchmarks have limited number of subjects (≤ 170) and modalities (≤ 2), which hinder the further development of the academic community. To facilitate face anti-spoofing research, we introduce a large-scale multi-modal dataset, namely CASIA-SURF, which is the largest publicly available dataset for face antispoofing in terms of both subjects and modalities. Specifically, it consists of 1, 000 subjects with 21, 000 videos and each sample has 3 modalities (i.e., RGB, Depth and IR). We also provide comprehensive evaluation metrics, diverse evaluation protocols, training/validation/testing subsets and a measurement tool, developing a new benchmark for face anti-spoofing. Moreover, we present a novel multi-modal multi-scale fusion method as a strong baseline, which performs feature re-weighting to select the more informative channel features while suppressing the less useful ones for each modality across different scales. Extensive experiments have been conducted on the proposed dataset to verify its significance and generalization capability. The dataset is available at https://sites.google.com/qq.com/chalearnfacespoofingattackdete/.

FeatherNets: Convolutional Neural Networks as Light as Feather for Face Anti-Spoofing

“…3)Recently, Zhang et al [9] provided a large-scale multimodal dataset, namely CASIA-SURF, which consists of 3 modalities data (RGB, depth and IR). It provides a strong baseline to make full use of these features by fusing multimodal data through a three-stream network.…”

“…ACER is treated as the evaluation metric, in which APCER and NPCER are used to measure the error rate of fake or real samples, respectively. Besides, the other metrics [9] are also used, such as TPR@FPR=10E-2, 10E-3, 10E-4.…”

“…Two datasets are used in the experiments: CASIA-SURF [9] and the proposed Multi-Modal Face Dataset (MMFD). CASIA-SURF is the largest publicly available dataset for face Anti-spoofing, provided by Surfing Technology [9]. It consists of 1,000 subjects with 21,000 videos and each sample has 3 modalities (i.e., RGB, Depth and IR), as shown in Figure 5.…”

“…Model ACER TPR@FPR=10E-2 TPR@FPR=10E-3 Params FLOPS ResNet18 [9] 0.05 0.883 0.272 11.18M 1800M Baseline [9] 0.0213 0.…”

“…We also design a new fusion classifier architecture which assembles and cascades several models learned from multimodal data, i.e., the depth and IR data, to generate better prediction accuracy than single depth models. Although the depth image is discriminative on 2D presentation attack detection, multi-modal fusion can boost the performance further due to its complementary and generalization capability [9]. The new fusion procedure has been applied to face anti-spoofing competition@CVPR2019 and showed the result of 0.0013 (ACER), 0.999 (TPR@FPR=10e-2), 0.998 (TPR@FPR=10e-3) and 0.9814 (TPR@FPR=10e-4) in the test dataset.…”