Matches in SemOpenAlex for { <https://semopenalex.org/work/W4312710588> ?p ?o ?g. }
- W4312710588 endingPage "94" @default.
- W4312710588 startingPage "76" @default.
- W4312710588 abstract "While recent advances in generative models benefit the society, the generated images can be abused for malicious purposes, like fraud, defamation, and false news. To prevent such cases, vigorous research is conducted on distinguishing the generated images from the real ones, but challenges still remain with detecting the unseen generated images outside of the training settings. To overcome this problem, we analyze the distinctive characteristic of the generated images called ‘fingerprints,’ and propose a new framework to reproduce diverse types of fingerprints generated by various generative models. By training the model with the real images only, our framework can avoid data dependency on particular generative models and enhance generalization. With the mathematical derivation that the fingerprint is emphasized at the frequency domain, we design a generated image detector for effective training of the fingerprints. Our framework outperforms the prior state-of-the-art detectors, even though only real images are used for training. We also provide new benchmark datasets to demonstrate the model’s robustness using the images of the latest anti-artifact generative models for reducing the spectral discrepancies." @default.
- W4312710588 created "2023-01-05" @default.
- W4312710588 creator A5003196471 @default.
- W4312710588 creator A5006852919 @default.
- W4312710588 creator A5012345993 @default.
- W4312710588 creator A5026297604 @default.
- W4312710588 creator A5078628073 @default.
- W4312710588 date "2022-01-01" @default.
- W4312710588 modified "2023-10-17" @default.
- W4312710588 title "FingerprintNet: Synthesized Fingerprints for Generated Image Detection" @default.
- W4312710588 cites W1834627138 @default.
- W4312710588 cites W1861492603 @default.
- W4312710588 cites W1975528596 @default.
- W4312710588 cites W2031489346 @default.
- W4312710588 cites W2031614119 @default.
- W4312710588 cites W2071794886 @default.
- W4312710588 cites W2096076665 @default.
- W4312710588 cites W2111374353 @default.
- W4312710588 cites W2117539524 @default.
- W4312710588 cites W2135461721 @default.
- W4312710588 cites W2489165573 @default.
- W4312710588 cites W2907295878 @default.
- W4312710588 cites W2911605501 @default.
- W4312710588 cites W2962770929 @default.
- W4312710588 cites W2962793481 @default.
- W4312710588 cites W2962974533 @default.
- W4312710588 cites W2963767194 @default.
- W4312710588 cites W3012472557 @default.
- W4312710588 cites W3034521057 @default.
- W4312710588 cites W3034577585 @default.
- W4312710588 cites W3034600949 @default.
- W4312710588 cites W3034625979 @default.
- W4312710588 cites W3034864980 @default.
- W4312710588 cites W3035574324 @default.
- W4312710588 cites W3102061158 @default.
- W4312710588 cites W3135133133 @default.
- W4312710588 cites W3158238516 @default.
- W4312710588 cites W3158353280 @default.
- W4312710588 cites W3166466426 @default.
- W4312710588 cites W3173126908 @default.
- W4312710588 cites W3173275980 @default.
- W4312710588 cites W3173317327 @default.
- W4312710588 cites W3174508664 @default.
- W4312710588 cites W3176241004 @default.
- W4312710588 cites W3177027279 @default.
- W4312710588 cites W3188897163 @default.
- W4312710588 cites W3191805365 @default.
- W4312710588 cites W3196551054 @default.
- W4312710588 cites W4214684483 @default.
- W4312710588 cites W4221149434 @default.
- W4312710588 doi "https://doi.org/10.1007/978-3-031-19781-9_5" @default.
- W4312710588 hasPublicationYear "2022" @default.
- W4312710588 type Work @default.
- W4312710588 citedByCount "2" @default.
- W4312710588 countsByYear W43127105882023 @default.
- W4312710588 crossrefType "book-chapter" @default.
- W4312710588 hasAuthorship W4312710588A5003196471 @default.
- W4312710588 hasAuthorship W4312710588A5006852919 @default.
- W4312710588 hasAuthorship W4312710588A5012345993 @default.
- W4312710588 hasAuthorship W4312710588A5026297604 @default.
- W4312710588 hasAuthorship W4312710588A5078628073 @default.
- W4312710588 hasConcept C104317684 @default.
- W4312710588 hasConcept C115961682 @default.
- W4312710588 hasConcept C119857082 @default.
- W4312710588 hasConcept C13280743 @default.
- W4312710588 hasConcept C134306372 @default.
- W4312710588 hasConcept C153180895 @default.
- W4312710588 hasConcept C154945302 @default.
- W4312710588 hasConcept C167966045 @default.
- W4312710588 hasConcept C177148314 @default.
- W4312710588 hasConcept C185592680 @default.
- W4312710588 hasConcept C185798385 @default.
- W4312710588 hasConcept C19768560 @default.
- W4312710588 hasConcept C205649164 @default.
- W4312710588 hasConcept C2777826928 @default.
- W4312710588 hasConcept C2779010991 @default.
- W4312710588 hasConcept C31972630 @default.
- W4312710588 hasConcept C33923547 @default.
- W4312710588 hasConcept C39890363 @default.
- W4312710588 hasConcept C41008148 @default.
- W4312710588 hasConcept C55493867 @default.
- W4312710588 hasConcept C63479239 @default.
- W4312710588 hasConceptScore W4312710588C104317684 @default.
- W4312710588 hasConceptScore W4312710588C115961682 @default.
- W4312710588 hasConceptScore W4312710588C119857082 @default.
- W4312710588 hasConceptScore W4312710588C13280743 @default.
- W4312710588 hasConceptScore W4312710588C134306372 @default.
- W4312710588 hasConceptScore W4312710588C153180895 @default.
- W4312710588 hasConceptScore W4312710588C154945302 @default.
- W4312710588 hasConceptScore W4312710588C167966045 @default.
- W4312710588 hasConceptScore W4312710588C177148314 @default.
- W4312710588 hasConceptScore W4312710588C185592680 @default.
- W4312710588 hasConceptScore W4312710588C185798385 @default.
- W4312710588 hasConceptScore W4312710588C19768560 @default.
- W4312710588 hasConceptScore W4312710588C205649164 @default.
- W4312710588 hasConceptScore W4312710588C2777826928 @default.
- W4312710588 hasConceptScore W4312710588C2779010991 @default.
- W4312710588 hasConceptScore W4312710588C31972630 @default.