Matches in SemOpenAlex for { <https://semopenalex.org/work/W4306315173> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W4306315173 abstract "A large number of deep neural network based techniques have been developed to address the challenging problem of face presentation attack detection (PAD). Whereas such techniques' focus has been on improving PAD performance in terms of classification accuracy and robustness against unseen attacks and environmental conditions, there exists little attention on the explainability of PAD predictions. In this paper, we tackle the problem of explaining PAD predictions through natural language. Our approach passes feature representations of a deep layer of the PAD model to a language model to generate text describing the reasoning behind the PAD prediction. Due to the limited amount of annotated data in our study, we apply a light-weight LSTM network as our natural language generation model. We investigate how the quality of the generated explanations is affected by different loss functions, including the commonly used word-wise cross entropy loss, a sentence discriminative loss, and a sentence semantic loss. We perform our experiments using face images from a dataset consisting of 1,105 bona-fide and 924 presentation attack samples. Our quantitative and qualitative results show the effectiveness of our model for generating proper PAD explanations through text as well as the power of the sentence-wise losses. To the best of our knowledge, this is the first introduction of a joint biometrics-NLP task. Our dataset can be obtained through our GitHub page." @default.
- W4306315173 created "2022-10-16" @default.
- W4306315173 creator A5000874697 @default.
- W4306315173 creator A5028776484 @default.
- W4306315173 creator A5070961790 @default.
- W4306315173 creator A5078745193 @default.
- W4306315173 creator A5083569557 @default.
- W4306315173 date "2021-11-08" @default.
- W4306315173 modified "2023-09-27" @default.
- W4306315173 title "Explaining Face Presentation Attack Detection Using Natural Language" @default.
- W4306315173 doi "https://doi.org/10.48550/arxiv.2111.04862" @default.
- W4306315173 hasPublicationYear "2021" @default.
- W4306315173 type Work @default.
- W4306315173 citedByCount "0" @default.
- W4306315173 crossrefType "posted-content" @default.
- W4306315173 hasAuthorship W4306315173A5000874697 @default.
- W4306315173 hasAuthorship W4306315173A5028776484 @default.
- W4306315173 hasAuthorship W4306315173A5070961790 @default.
- W4306315173 hasAuthorship W4306315173A5078745193 @default.
- W4306315173 hasAuthorship W4306315173A5083569557 @default.
- W4306315173 hasBestOaLocation W43063151731 @default.
- W4306315173 hasConcept C104317684 @default.
- W4306315173 hasConcept C108583219 @default.
- W4306315173 hasConcept C138885662 @default.
- W4306315173 hasConcept C153180895 @default.
- W4306315173 hasConcept C154945302 @default.
- W4306315173 hasConcept C167981619 @default.
- W4306315173 hasConcept C184297639 @default.
- W4306315173 hasConcept C185592680 @default.
- W4306315173 hasConcept C188441871 @default.
- W4306315173 hasConcept C195324797 @default.
- W4306315173 hasConcept C204321447 @default.
- W4306315173 hasConcept C2776401178 @default.
- W4306315173 hasConcept C2777530160 @default.
- W4306315173 hasConcept C2779304628 @default.
- W4306315173 hasConcept C28490314 @default.
- W4306315173 hasConcept C41008148 @default.
- W4306315173 hasConcept C41895202 @default.
- W4306315173 hasConcept C55493867 @default.
- W4306315173 hasConcept C63479239 @default.
- W4306315173 hasConcept C97931131 @default.
- W4306315173 hasConceptScore W4306315173C104317684 @default.
- W4306315173 hasConceptScore W4306315173C108583219 @default.
- W4306315173 hasConceptScore W4306315173C138885662 @default.
- W4306315173 hasConceptScore W4306315173C153180895 @default.
- W4306315173 hasConceptScore W4306315173C154945302 @default.
- W4306315173 hasConceptScore W4306315173C167981619 @default.
- W4306315173 hasConceptScore W4306315173C184297639 @default.
- W4306315173 hasConceptScore W4306315173C185592680 @default.
- W4306315173 hasConceptScore W4306315173C188441871 @default.
- W4306315173 hasConceptScore W4306315173C195324797 @default.
- W4306315173 hasConceptScore W4306315173C204321447 @default.
- W4306315173 hasConceptScore W4306315173C2776401178 @default.
- W4306315173 hasConceptScore W4306315173C2777530160 @default.
- W4306315173 hasConceptScore W4306315173C2779304628 @default.
- W4306315173 hasConceptScore W4306315173C28490314 @default.
- W4306315173 hasConceptScore W4306315173C41008148 @default.
- W4306315173 hasConceptScore W4306315173C41895202 @default.
- W4306315173 hasConceptScore W4306315173C55493867 @default.
- W4306315173 hasConceptScore W4306315173C63479239 @default.
- W4306315173 hasConceptScore W4306315173C97931131 @default.
- W4306315173 hasLocation W43063151731 @default.
- W4306315173 hasOpenAccess W4306315173 @default.
- W4306315173 hasPrimaryLocation W43063151731 @default.
- W4306315173 hasRelatedWork W1567338489 @default.
- W4306315173 hasRelatedWork W159132833 @default.
- W4306315173 hasRelatedWork W2593180407 @default.
- W4306315173 hasRelatedWork W2743258233 @default.
- W4306315173 hasRelatedWork W2766123424 @default.
- W4306315173 hasRelatedWork W2782592381 @default.
- W4306315173 hasRelatedWork W3213462841 @default.
- W4306315173 hasRelatedWork W4226341954 @default.
- W4306315173 hasRelatedWork W4323060069 @default.
- W4306315173 hasRelatedWork W4379256022 @default.
- W4306315173 isParatext "false" @default.
- W4306315173 isRetracted "false" @default.
- W4306315173 workType "article" @default.