FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4210595036> ?p ?o ?g. }

• W4210595036 endingPage "100217" @default.
• W4210595036 startingPage "100217" @default.
• W4210595036 abstract "Deepfakes have become exponentially more common and sophisticated in recent years, so much so that forensic specialists, policy makers, and the public alike are anxious about their role in spreading disinformation. Recently, the detection and creation of such forgery became a popular research topic, leading to significant growth in publications related to the creation of deepfakes, detection methods, and datasets containing the latest deepfake creation methods. The most successful approaches in identifying and preventing deepfakes are deep learning methods that rely on convolutional neural networks as the backbone for a binary classification task. A convolutional neural network extracts the underlying patterns from the input frames. It feeds these to a binary classification fully connected network, which classifies these patterns as trustworthy or untrustworthy. We claim that this method is not ideal in a scenario in which the generation algorithms constantly evolve since the detection algorithm is not robust enough to detect comparably minor artifacts introduced by the generation algorithms. This work proposes a hierarchical explainable forensics algorithm that incorporates humans in the detection loop. We curate the data through a deep learning detection algorithm and share an explainable decision to humans alongside a set of forensic analyses on the decision region. On the detection side, we propose an attention-based explainable deepfake detection algorithm. We address this generalization issue by implementing an ensemble of standard and attention-based data-augmented detection networks. We use the attention blocks to evaluate the face regions where the model focuses its decision. We simultaneously drop and enlarge the region to push the model to base its decision on more regions of the face, while maintaining a specific focal point for its decision. In this case, we use an ensemble of models to improve the generalization. We also evaluate the models' decision using Grad-CAM explanation to focus on the attention maps. The region uncovered by the explanation layer is cropped and undergoes a series of frequency and statistical analyses that help humans decide if the frame is real or fake. We evaluate our model in one of the most challenging datasets, the DFDC, and achieve an accuracy of 92.4%. We successfully maintain this accuracy in datasets not used in the training process." @default.
• W4210595036 created "2022-02-08" @default.
• W4210595036 creator A5011800243 @default.
• W4210595036 creator A5012889978 @default.
• W4210595036 creator A5013922720 @default.
• W4210595036 creator A5017210698 @default.
• W4210595036 creator A5035502033 @default.
• W4210595036 creator A5063952127 @default.
• W4210595036 date "2022-01-01" @default.
• W4210595036 modified "2023-10-14" @default.
• W4210595036 title "Deepfake forensics analysis: An explainable hierarchical ensemble of weakly supervised models" @default.
• W4210595036 cites W2146991130 @default.
• W4210595036 cites W2534699224 @default.
• W4210595036 cites W2920877282 @default.
• W4210595036 cites W2963881378 @default.
• W4210595036 cites W2970868842 @default.
• W4210595036 cites W2989851933 @default.
• W4210595036 cites W3013122344 @default.
• W4210595036 cites W3038930935 @default.
• W4210595036 cites W3125803510 @default.
• W4210595036 doi "https://doi.org/10.1016/j.fsisyn.2022.100217" @default.
• W4210595036 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/35128371" @default.
• W4210595036 hasPublicationYear "2022" @default.
• W4210595036 type Work @default.
• W4210595036 citedByCount "5" @default.
• W4210595036 countsByYear W42105950362022 @default.
• W4210595036 crossrefType "journal-article" @default.
• W4210595036 hasAuthorship W4210595036A5011800243 @default.
• W4210595036 hasAuthorship W4210595036A5012889978 @default.
• W4210595036 hasAuthorship W4210595036A5013922720 @default.
• W4210595036 hasAuthorship W4210595036A5017210698 @default.
• W4210595036 hasAuthorship W4210595036A5035502033 @default.
• W4210595036 hasAuthorship W4210595036A5063952127 @default.
• W4210595036 hasBestOaLocation W42105950361 @default.
• W4210595036 hasConcept C108583219 @default.
• W4210595036 hasConcept C119857082 @default.
• W4210595036 hasConcept C124101348 @default.
• W4210595036 hasConcept C134306372 @default.
• W4210595036 hasConcept C154945302 @default.
• W4210595036 hasConcept C177148314 @default.
• W4210595036 hasConcept C177264268 @default.
• W4210595036 hasConcept C199360897 @default.
• W4210595036 hasConcept C33923547 @default.
• W4210595036 hasConcept C41008148 @default.
• W4210595036 hasConcept C45942800 @default.
• W4210595036 hasConcept C81363708 @default.
• W4210595036 hasConceptScore W4210595036C108583219 @default.
• W4210595036 hasConceptScore W4210595036C119857082 @default.
• W4210595036 hasConceptScore W4210595036C124101348 @default.
• W4210595036 hasConceptScore W4210595036C134306372 @default.
• W4210595036 hasConceptScore W4210595036C154945302 @default.
• W4210595036 hasConceptScore W4210595036C177148314 @default.
• W4210595036 hasConceptScore W4210595036C177264268 @default.
• W4210595036 hasConceptScore W4210595036C199360897 @default.
• W4210595036 hasConceptScore W4210595036C33923547 @default.
• W4210595036 hasConceptScore W4210595036C41008148 @default.
• W4210595036 hasConceptScore W4210595036C45942800 @default.
• W4210595036 hasConceptScore W4210595036C81363708 @default.
• W4210595036 hasLocation W42105950361 @default.
• W4210595036 hasLocation W42105950362 @default.
• W4210595036 hasLocation W42105950363 @default.
• W4210595036 hasLocation W42105950364 @default.
• W4210595036 hasOpenAccess W4210595036 @default.
• W4210595036 hasPrimaryLocation W42105950361 @default.
• W4210595036 hasRelatedWork W2731899572 @default.
• W4210595036 hasRelatedWork W2810053714 @default.
• W4210595036 hasRelatedWork W2999805992 @default.
• W4210595036 hasRelatedWork W3116150086 @default.
• W4210595036 hasRelatedWork W3133861977 @default.
• W4210595036 hasRelatedWork W3136979370 @default.
• W4210595036 hasRelatedWork W4200173597 @default.
• W4210595036 hasRelatedWork W4312417841 @default.
• W4210595036 hasRelatedWork W4321369474 @default.
• W4210595036 hasRelatedWork W4380075502 @default.
• W4210595036 hasVolume "4" @default.
• W4210595036 isParatext "false" @default.
• W4210595036 isRetracted "false" @default.
• W4210595036 workType "article" @default.