Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386043282> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W4386043282 abstract "Learning generalized face anti-spoofing (FAS) models against presentation attacks is essential for the security of face recognition systems. Previous FAS methods usually encourage models to extract discriminative features, of which the distances within the same class (bonafide or attack) are pushed close while those between bonafide and attack are pulled away. However, these methods are designed based on Euclidean distance, which lacks generalization ability for unseen attack detection due to poor hierarchy embedding ability. According to the evidence that different spoofing attacks are intrinsically hierarchical, we propose to learn richer hierarchical and discriminative spoofing cues in hyperbolic space. Specifically, for unimodal FAS learning, the feature embeddings are projected into the Poincar'e ball, and then the hyperbolic binary logistic regression layer is cascaded for classification. To further improve generalization, we conduct hyperbolic contrastive learning for the bonafide only while relaxing the constraints on diverse spoofing attacks. To alleviate the vanishing gradient problem in hyperbolic space, a new feature clipping method is proposed to enhance the training stability of hyperbolic models. Besides, we further design a multimodal FAS framework with Euclidean multimodal feature decomposition and hyperbolic multimodal feature fusion & classification. Extensive experiments on three benchmark datasets (i.e., WMCA, PADISI-Face, and SiW-M) with diverse attack types demonstrate that the proposed method can bring significant improvement compared to the Euclidean baselines on unseen attack detection. In addition, the proposed framework is also generalized well on four benchmark datasets (i.e., MSU-MFSD, IDIAP REPLAY-ATTACK, CASIA-FASD, and OULU-NPU) with a limited number of attack types." @default.
- W4386043282 created "2023-08-22" @default.
- W4386043282 creator A5010576164 @default.
- W4386043282 creator A5016271966 @default.
- W4386043282 creator A5040232431 @default.
- W4386043282 creator A5043603162 @default.
- W4386043282 creator A5062522283 @default.
- W4386043282 creator A5068658382 @default.
- W4386043282 date "2023-08-17" @default.
- W4386043282 modified "2023-09-25" @default.
- W4386043282 title "Hyperbolic Face Anti-Spoofing" @default.
- W4386043282 doi "https://doi.org/10.48550/arxiv.2308.09107" @default.
- W4386043282 hasPublicationYear "2023" @default.
- W4386043282 type Work @default.
- W4386043282 citedByCount "0" @default.
- W4386043282 crossrefType "posted-content" @default.
- W4386043282 hasAuthorship W4386043282A5010576164 @default.
- W4386043282 hasAuthorship W4386043282A5016271966 @default.
- W4386043282 hasAuthorship W4386043282A5040232431 @default.
- W4386043282 hasAuthorship W4386043282A5043603162 @default.
- W4386043282 hasAuthorship W4386043282A5062522283 @default.
- W4386043282 hasAuthorship W4386043282A5068658382 @default.
- W4386043282 hasBestOaLocation W43860432821 @default.
- W4386043282 hasConcept C134306372 @default.
- W4386043282 hasConcept C138885662 @default.
- W4386043282 hasConcept C144024400 @default.
- W4386043282 hasConcept C153180895 @default.
- W4386043282 hasConcept C154945302 @default.
- W4386043282 hasConcept C167900197 @default.
- W4386043282 hasConcept C177148314 @default.
- W4386043282 hasConcept C202444582 @default.
- W4386043282 hasConcept C2776401178 @default.
- W4386043282 hasConcept C2779304628 @default.
- W4386043282 hasConcept C33923547 @default.
- W4386043282 hasConcept C36289849 @default.
- W4386043282 hasConcept C38652104 @default.
- W4386043282 hasConcept C41008148 @default.
- W4386043282 hasConcept C41608201 @default.
- W4386043282 hasConcept C41895202 @default.
- W4386043282 hasConcept C52622490 @default.
- W4386043282 hasConcept C83665646 @default.
- W4386043282 hasConcept C83677898 @default.
- W4386043282 hasConcept C97931131 @default.
- W4386043282 hasConceptScore W4386043282C134306372 @default.
- W4386043282 hasConceptScore W4386043282C138885662 @default.
- W4386043282 hasConceptScore W4386043282C144024400 @default.
- W4386043282 hasConceptScore W4386043282C153180895 @default.
- W4386043282 hasConceptScore W4386043282C154945302 @default.
- W4386043282 hasConceptScore W4386043282C167900197 @default.
- W4386043282 hasConceptScore W4386043282C177148314 @default.
- W4386043282 hasConceptScore W4386043282C202444582 @default.
- W4386043282 hasConceptScore W4386043282C2776401178 @default.
- W4386043282 hasConceptScore W4386043282C2779304628 @default.
- W4386043282 hasConceptScore W4386043282C33923547 @default.
- W4386043282 hasConceptScore W4386043282C36289849 @default.
- W4386043282 hasConceptScore W4386043282C38652104 @default.
- W4386043282 hasConceptScore W4386043282C41008148 @default.
- W4386043282 hasConceptScore W4386043282C41608201 @default.
- W4386043282 hasConceptScore W4386043282C41895202 @default.
- W4386043282 hasConceptScore W4386043282C52622490 @default.
- W4386043282 hasConceptScore W4386043282C83665646 @default.
- W4386043282 hasConceptScore W4386043282C83677898 @default.
- W4386043282 hasConceptScore W4386043282C97931131 @default.
- W4386043282 hasLocation W43860432821 @default.
- W4386043282 hasOpenAccess W4386043282 @default.
- W4386043282 hasPrimaryLocation W43860432821 @default.
- W4386043282 hasRelatedWork W1461621550 @default.
- W4386043282 hasRelatedWork W1775397219 @default.
- W4386043282 hasRelatedWork W2020350089 @default.
- W4386043282 hasRelatedWork W2285052147 @default.
- W4386043282 hasRelatedWork W2546942002 @default.
- W4386043282 hasRelatedWork W2772780115 @default.
- W4386043282 hasRelatedWork W2773500201 @default.
- W4386043282 hasRelatedWork W2806866760 @default.
- W4386043282 hasRelatedWork W2944661354 @default.
- W4386043282 hasRelatedWork W2970216048 @default.
- W4386043282 isParatext "false" @default.
- W4386043282 isRetracted "false" @default.
- W4386043282 workType "article" @default.