Matches in SemOpenAlex for { <https://semopenalex.org/work/W4304080478> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W4304080478 abstract "With various unforeseeable face presentation attacks (PA) springing up, face anti-spoofing (FAS) urgently needs to generalize to unseen scenarios. Research on generalizable FAS has lately attracted growing attention. Existing methods cast FAS as a vanilla binary classification problem and address it by a standard discriminative classifier p(y|x) under a domain generalization framework. However, discriminative models are unreliable for samples far away from the training distribution. In this paper, we resort to an energy-based model (EBM) to tackle FAS in a generative perspective. Our motivation is to model the joint density p(x,y), which allows to compute not only p(y|x) but also p(x). Due to the intractability of direct modeling, we use EBMs as an alternative to probabilistic estimation. With energy-based training, real faces are encouraged to get low free energy associated with the marginal probability p(x) of real faces, and all samples with high free energy are regarded as fake faces, thus rejecting any kind of PA out of the distribution of real faces. To learn to generalize to unseen domains, we generate diverse and novel populations in feature space under the guidance of energy model. Our model is updated in a meta-learning schema, where the original source samples are utilized for meta-training and the generated ones for meta-testing. We validate our method on four widely used FAS datasets. Comprehensive experimental results demonstrate the effectiveness of our method compared with state-of-the-arts." @default.
- W4304080478 created "2022-10-10" @default.
- W4304080478 creator A5003786587 @default.
- W4304080478 creator A5013617878 @default.
- W4304080478 creator A5057820981 @default.
- W4304080478 creator A5079281855 @default.
- W4304080478 creator A5089989183 @default.
- W4304080478 date "2022-10-10" @default.
- W4304080478 modified "2023-09-26" @default.
- W4304080478 title "Energy-Based Domain Generalization for Face Anti-Spoofing" @default.
- W4304080478 cites W1835705677 @default.
- W4304080478 cites W1989702938 @default.
- W4304080478 cites W1994030682 @default.
- W4304080478 cites W2042883034 @default.
- W4304080478 cites W2095252718 @default.
- W4304080478 cites W2118392996 @default.
- W4304080478 cites W2174309130 @default.
- W4304080478 cites W2728977829 @default.
- W4304080478 cites W2763549966 @default.
- W4304080478 cites W2952476201 @default.
- W4304080478 cites W2955273087 @default.
- W4304080478 cites W2956066883 @default.
- W4304080478 cites W2963342110 @default.
- W4304080478 cites W2974497444 @default.
- W4304080478 cites W2981446661 @default.
- W4304080478 cites W2998163473 @default.
- W4304080478 cites W3006683596 @default.
- W4304080478 cites W3018638193 @default.
- W4304080478 cites W3034373371 @default.
- W4304080478 cites W3034594921 @default.
- W4304080478 cites W3035436173 @default.
- W4304080478 cites W3159891475 @default.
- W4304080478 cites W3181127262 @default.
- W4304080478 cites W3198731329 @default.
- W4304080478 cites W3205314902 @default.
- W4304080478 cites W3205799947 @default.
- W4304080478 doi "https://doi.org/10.1145/3503161.3548073" @default.
- W4304080478 hasPublicationYear "2022" @default.
- W4304080478 type Work @default.
- W4304080478 citedByCount "1" @default.
- W4304080478 countsByYear W43040804782023 @default.
- W4304080478 crossrefType "proceedings-article" @default.
- W4304080478 hasAuthorship W4304080478A5003786587 @default.
- W4304080478 hasAuthorship W4304080478A5013617878 @default.
- W4304080478 hasAuthorship W4304080478A5057820981 @default.
- W4304080478 hasAuthorship W4304080478A5079281855 @default.
- W4304080478 hasAuthorship W4304080478A5089989183 @default.
- W4304080478 hasConcept C107673813 @default.
- W4304080478 hasConcept C114289077 @default.
- W4304080478 hasConcept C119857082 @default.
- W4304080478 hasConcept C134306372 @default.
- W4304080478 hasConcept C153180895 @default.
- W4304080478 hasConcept C154945302 @default.
- W4304080478 hasConcept C177148314 @default.
- W4304080478 hasConcept C33923547 @default.
- W4304080478 hasConcept C41008148 @default.
- W4304080478 hasConcept C49937458 @default.
- W4304080478 hasConcept C83665646 @default.
- W4304080478 hasConcept C95623464 @default.
- W4304080478 hasConcept C97931131 @default.
- W4304080478 hasConceptScore W4304080478C107673813 @default.
- W4304080478 hasConceptScore W4304080478C114289077 @default.
- W4304080478 hasConceptScore W4304080478C119857082 @default.
- W4304080478 hasConceptScore W4304080478C134306372 @default.
- W4304080478 hasConceptScore W4304080478C153180895 @default.
- W4304080478 hasConceptScore W4304080478C154945302 @default.
- W4304080478 hasConceptScore W4304080478C177148314 @default.
- W4304080478 hasConceptScore W4304080478C33923547 @default.
- W4304080478 hasConceptScore W4304080478C41008148 @default.
- W4304080478 hasConceptScore W4304080478C49937458 @default.
- W4304080478 hasConceptScore W4304080478C83665646 @default.
- W4304080478 hasConceptScore W4304080478C95623464 @default.
- W4304080478 hasConceptScore W4304080478C97931131 @default.
- W4304080478 hasFunder F4320321001 @default.
- W4304080478 hasLocation W43040804781 @default.
- W4304080478 hasOpenAccess W4304080478 @default.
- W4304080478 hasPrimaryLocation W43040804781 @default.
- W4304080478 hasRelatedWork W1461621550 @default.
- W4304080478 hasRelatedWork W1914651075 @default.
- W4304080478 hasRelatedWork W2020350089 @default.
- W4304080478 hasRelatedWork W2112343299 @default.
- W4304080478 hasRelatedWork W2143784992 @default.
- W4304080478 hasRelatedWork W2184708910 @default.
- W4304080478 hasRelatedWork W2311479510 @default.
- W4304080478 hasRelatedWork W2773500201 @default.
- W4304080478 hasRelatedWork W2905846897 @default.
- W4304080478 hasRelatedWork W62482549 @default.
- W4304080478 isParatext "false" @default.
- W4304080478 isRetracted "false" @default.
- W4304080478 workType "article" @default.