Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385800904> ?p ?o ?g. }
- W4385800904 abstract "Face anti-spoofing (FAS) is indispensable for a face recognition system. Many texture-driven countermeasures were developed against presentation attacks (PAs), but the performance against unseen domains or unseen spoofing types is still unsatisfactory. Instead of exhaustively collecting all the spoofing variations and making binary decisions of live/spoof, we offer a new perspective on the FAS task to distinguish between normal and abnormal movements of live and spoof presentations. We propose Geometry-Aware Interaction Network (GAIN), which exploits dense facial landmarks with spatio-temporal graph convolutional network (ST-GCN) to establish a more interpretable and modularized FAS model. Additionally, with our cross-attention feature interaction mechanism, GAIN can be easily integrated with other existing methods to significantly boost performance. Our approach achieves state-of-the-art performance in the standard intra- and cross-dataset evaluations. Moreover, our model outperforms state-of-the-art methods by a large margin in the cross-dataset cross-type protocol on CASIA-SURF 3DMask (+10.26 higher AUC score), exhibiting strong robustness against domain shifts and unseen spoofing types." @default.
- W4385800904 created "2023-08-15" @default.
- W4385800904 creator A5043359967 @default.
- W4385800904 creator A5054702628 @default.
- W4385800904 creator A5073849580 @default.
- W4385800904 creator A5074346604 @default.
- W4385800904 creator A5075373716 @default.
- W4385800904 creator A5086485553 @default.
- W4385800904 creator A5088397557 @default.
- W4385800904 date "2023-06-01" @default.
- W4385800904 modified "2023-09-26" @default.
- W4385800904 title "A Closer Look at Geometric Temporal Dynamics for Face Anti-Spoofing" @default.
- W4385800904 cites W1982209341 @default.
- W4385800904 cites W2003092530 @default.
- W4385800904 cites W2194775991 @default.
- W4385800904 cites W2265959009 @default.
- W4385800904 cites W2510185399 @default.
- W4385800904 cites W2612707971 @default.
- W4385800904 cites W2728977829 @default.
- W4385800904 cites W2787613668 @default.
- W4385800904 cites W2790392345 @default.
- W4385800904 cites W2948058585 @default.
- W4385800904 cites W2949699261 @default.
- W4385800904 cites W2952476201 @default.
- W4385800904 cites W2952587893 @default.
- W4385800904 cites W2956066883 @default.
- W4385800904 cites W2962934715 @default.
- W4385800904 cites W2963076818 @default.
- W4385800904 cites W2963656031 @default.
- W4385800904 cites W2964003763 @default.
- W4385800904 cites W2964134613 @default.
- W4385800904 cites W2979481854 @default.
- W4385800904 cites W3005726865 @default.
- W4385800904 cites W3034594921 @default.
- W4385800904 cites W3035116217 @default.
- W4385800904 cites W3035225512 @default.
- W4385800904 cites W3035263140 @default.
- W4385800904 cites W3035459165 @default.
- W4385800904 cites W3094861582 @default.
- W4385800904 cites W3098538019 @default.
- W4385800904 cites W3104792420 @default.
- W4385800904 cites W3109432287 @default.
- W4385800904 cites W3183833050 @default.
- W4385800904 cites W3188354902 @default.
- W4385800904 cites W3191040240 @default.
- W4385800904 cites W3199885658 @default.
- W4385800904 cites W3203634062 @default.
- W4385800904 cites W3204715535 @default.
- W4385800904 cites W3205638629 @default.
- W4385800904 cites W3207485660 @default.
- W4385800904 cites W3210533571 @default.
- W4385800904 cites W3212258096 @default.
- W4385800904 cites W4226291055 @default.
- W4385800904 cites W4285345751 @default.
- W4385800904 cites W4304092679 @default.
- W4385800904 cites W4312836647 @default.
- W4385800904 doi "https://doi.org/10.1109/cvprw59228.2023.00115" @default.
- W4385800904 hasPublicationYear "2023" @default.
- W4385800904 type Work @default.
- W4385800904 citedByCount "0" @default.
- W4385800904 crossrefType "proceedings-article" @default.
- W4385800904 hasAuthorship W4385800904A5043359967 @default.
- W4385800904 hasAuthorship W4385800904A5054702628 @default.
- W4385800904 hasAuthorship W4385800904A5073849580 @default.
- W4385800904 hasAuthorship W4385800904A5074346604 @default.
- W4385800904 hasAuthorship W4385800904A5075373716 @default.
- W4385800904 hasAuthorship W4385800904A5086485553 @default.
- W4385800904 hasAuthorship W4385800904A5088397557 @default.
- W4385800904 hasBestOaLocation W43858009042 @default.
- W4385800904 hasConcept C104317684 @default.
- W4385800904 hasConcept C115961682 @default.
- W4385800904 hasConcept C119857082 @default.
- W4385800904 hasConcept C132525143 @default.
- W4385800904 hasConcept C138885662 @default.
- W4385800904 hasConcept C144024400 @default.
- W4385800904 hasConcept C153180895 @default.
- W4385800904 hasConcept C154945302 @default.
- W4385800904 hasConcept C167900197 @default.
- W4385800904 hasConcept C185592680 @default.
- W4385800904 hasConcept C2776401178 @default.
- W4385800904 hasConcept C2779304628 @default.
- W4385800904 hasConcept C31510193 @default.
- W4385800904 hasConcept C31972630 @default.
- W4385800904 hasConcept C33923547 @default.
- W4385800904 hasConcept C36289849 @default.
- W4385800904 hasConcept C38652104 @default.
- W4385800904 hasConcept C41008148 @default.
- W4385800904 hasConcept C41895202 @default.
- W4385800904 hasConcept C48372109 @default.
- W4385800904 hasConcept C52622490 @default.
- W4385800904 hasConcept C53533937 @default.
- W4385800904 hasConcept C55493867 @default.
- W4385800904 hasConcept C63479239 @default.
- W4385800904 hasConcept C63584917 @default.
- W4385800904 hasConcept C774472 @default.
- W4385800904 hasConcept C80444323 @default.
- W4385800904 hasConcept C81363708 @default.
- W4385800904 hasConcept C87335442 @default.
- W4385800904 hasConcept C94375191 @default.
- W4385800904 hasConceptScore W4385800904C104317684 @default.