Matches in SemOpenAlex for { <https://semopenalex.org/work/W3003340385> ?p ?o ?g. }
- W3003340385 abstract "Generative Adversarial Networks (GANs) can generate realistic fake face images that can easily fool human beings.On the contrary, a common Convolutional Neural Network(CNN) discriminator can achieve more than 99.9% accuracyin discerning fake/real images. In this paper, we conduct an empirical study on fake/real faces, and have two important observations: firstly, the texture of fake faces is substantially different from real ones; secondly, global texture statistics are more robust to image editing and transferable to fake faces from different GANs and datasets. Motivated by the above observations, we propose a new architecture coined as Gram-Net, which leverages global image texture representations for robust fake image detection. Experimental results on several datasets demonstrate that our Gram-Net outperforms existing approaches. Especially, our Gram-Netis more robust to image editings, e.g. down-sampling, JPEG compression, blur, and noise. More importantly, our Gram-Net generalizes significantly better in detecting fake faces from GAN models not seen in the training phase and can perform decently in detecting fake natural images." @default.
- W3003340385 created "2020-02-07" @default.
- W3003340385 creator A5014239420 @default.
- W3003340385 creator A5042899882 @default.
- W3003340385 creator A5048895819 @default.
- W3003340385 date "2020-01-31" @default.
- W3003340385 modified "2023-10-01" @default.
- W3003340385 title "Global Texture Enhancement for Fake Face Detection in the Wild" @default.
- W3003340385 cites W1834627138 @default.
- W3003340385 cites W1915485278 @default.
- W3003340385 cites W2044465660 @default.
- W3003340385 cites W2061052400 @default.
- W3003340385 cites W2099471712 @default.
- W3003340385 cites W2115252128 @default.
- W3003340385 cites W2163605009 @default.
- W3003340385 cites W2173520492 @default.
- W3003340385 cites W2194775991 @default.
- W3003340385 cites W2295107390 @default.
- W3003340385 cites W2475287302 @default.
- W3003340385 cites W2556967412 @default.
- W3003340385 cites W2605195953 @default.
- W3003340385 cites W2739748921 @default.
- W3003340385 cites W2775288145 @default.
- W3003340385 cites W2811414481 @default.
- W3003340385 cites W2888519208 @default.
- W3003340385 cites W2893749619 @default.
- W3003340385 cites W2899771611 @default.
- W3003340385 cites W2907295878 @default.
- W3003340385 cites W2921416006 @default.
- W3003340385 cites W2952610664 @default.
- W3003340385 cites W2959409678 @default.
- W3003340385 cites W2962760235 @default.
- W3003340385 cites W2962770929 @default.
- W3003340385 cites W2962793481 @default.
- W3003340385 cites W2962879692 @default.
- W3003340385 cites W2962947361 @default.
- W3003340385 cites W2963767194 @default.
- W3003340385 cites W2964193438 @default.
- W3003340385 cites W2972754346 @default.
- W3003340385 cites W2978778164 @default.
- W3003340385 cites W2979980060 @default.
- W3003340385 cites W2981803355 @default.
- W3003340385 cites W3012472557 @default.
- W3003340385 doi "https://doi.org/10.48550/arxiv.2002.00133" @default.
- W3003340385 hasPublicationYear "2020" @default.
- W3003340385 type Work @default.
- W3003340385 sameAs 3003340385 @default.
- W3003340385 citedByCount "4" @default.
- W3003340385 countsByYear W30033403852020 @default.
- W3003340385 countsByYear W30033403852021 @default.
- W3003340385 crossrefType "posted-content" @default.
- W3003340385 hasAuthorship W3003340385A5014239420 @default.
- W3003340385 hasAuthorship W3003340385A5042899882 @default.
- W3003340385 hasAuthorship W3003340385A5048895819 @default.
- W3003340385 hasBestOaLocation W30033403851 @default.
- W3003340385 hasConcept C115961682 @default.
- W3003340385 hasConcept C144024400 @default.
- W3003340385 hasConcept C153180895 @default.
- W3003340385 hasConcept C154945302 @default.
- W3003340385 hasConcept C2776674983 @default.
- W3003340385 hasConcept C2779304628 @default.
- W3003340385 hasConcept C2779803651 @default.
- W3003340385 hasConcept C2781195486 @default.
- W3003340385 hasConcept C31972630 @default.
- W3003340385 hasConcept C36289849 @default.
- W3003340385 hasConcept C41008148 @default.
- W3003340385 hasConcept C76155785 @default.
- W3003340385 hasConcept C81363708 @default.
- W3003340385 hasConcept C94915269 @default.
- W3003340385 hasConcept C99498987 @default.
- W3003340385 hasConceptScore W3003340385C115961682 @default.
- W3003340385 hasConceptScore W3003340385C144024400 @default.
- W3003340385 hasConceptScore W3003340385C153180895 @default.
- W3003340385 hasConceptScore W3003340385C154945302 @default.
- W3003340385 hasConceptScore W3003340385C2776674983 @default.
- W3003340385 hasConceptScore W3003340385C2779304628 @default.
- W3003340385 hasConceptScore W3003340385C2779803651 @default.
- W3003340385 hasConceptScore W3003340385C2781195486 @default.
- W3003340385 hasConceptScore W3003340385C31972630 @default.
- W3003340385 hasConceptScore W3003340385C36289849 @default.
- W3003340385 hasConceptScore W3003340385C41008148 @default.
- W3003340385 hasConceptScore W3003340385C76155785 @default.
- W3003340385 hasConceptScore W3003340385C81363708 @default.
- W3003340385 hasConceptScore W3003340385C94915269 @default.
- W3003340385 hasConceptScore W3003340385C99498987 @default.
- W3003340385 hasLocation W30033403851 @default.
- W3003340385 hasLocation W30033403852 @default.
- W3003340385 hasOpenAccess W3003340385 @default.
- W3003340385 hasPrimaryLocation W30033403851 @default.
- W3003340385 hasRelatedWork W1899364738 @default.
- W3003340385 hasRelatedWork W2092957489 @default.
- W3003340385 hasRelatedWork W2139117675 @default.
- W3003340385 hasRelatedWork W2348423830 @default.
- W3003340385 hasRelatedWork W2366116130 @default.
- W3003340385 hasRelatedWork W2372194912 @default.
- W3003340385 hasRelatedWork W2521062615 @default.
- W3003340385 hasRelatedWork W2976316911 @default.
- W3003340385 hasRelatedWork W3093612317 @default.
- W3003340385 hasRelatedWork W43171467 @default.
- W3003340385 isParatext "false" @default.