Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Adaptive Transformers for Robust Few-shot Cross-domain Face Anti-spoofing

About

While recent face anti-spoofing methods perform well under the intra-domain setups, an effective approach needs to account for much larger appearance variations of images acquired in complex scenes with different sensors for robust performance. In this paper, we present adaptive vision transformers (ViT) for robust cross-domain face antispoofing. Specifically, we adopt ViT as a backbone to exploit its strength to account for long-range dependencies among pixels. We further introduce the ensemble adapters module and feature-wise transformation layers in the ViT to adapt to different domains for robust performance with a few samples. Experiments on several benchmark datasets show that the proposed models achieve both robust and competitive performance against the state-of-the-art methods for cross-domain face anti-spoofing using a few samples.

Hsin-Ping Huang, Deqing Sun, Yaojie Liu, Wen-Sheng Chu, Taihong Xiao, Jinwei Yuan, Hartwig Adam, Ming-Hsuan Yang• 2022

Related benchmarks

TaskDatasetResultRank
Face Anti-SpoofingCASIA-CeFA (C), PADISI (P), CASIA-SURF (S), and WMCA (W) Protocol 2, Missing D
HTER6.83
49
Face Anti-SpoofingReplay-Attack (test)
HTER12.38
38
Face Anti-SpoofingMSU-MFSD
HTER (%)12.86
18
Face Anti-SpoofingOULU-NPU
HTER26.73
18
Face Anti-SpoofingWMCA
HTER29.88
18
Face Anti-SpoofingCASIA-MFSD
HTER3.11
18
Face Anti-SpoofingHiFiMask
HTER (%)37.3
18
Face Anti-SpoofingSIW
HTER14.74
18
Face Anti-SpoofingSIW-M V2
HTER26.72
18
Face Anti-SpoofingCASIA-SURF-3DMask
HTER6.18
13
Showing 10 of 27 rows

Other info

Follow for update