FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3110828520> ?p ?o ?g. }

• W3110828520 abstract "Although current deep learning-based face forgery detectors achieve impressive performance in constrained scenarios, they are vulnerable to samples created by unseen manipulation methods. Some recent works show improvements in generalisation but rely on cues that are easily corrupted by common post-processing operations such as compression. In this paper, we propose LipForensics, a detection approach capable of both generalising to novel manipulations and withstanding various distortions. LipForensics targets high-level semantic irregularities in mouth movements, which are common in many generated videos. It consists in first pretraining a spatio-temporal network to perform visual speech recognition (lipreading), thus learning rich internal representations related to natural mouth motion. A temporal network is subsequently finetuned on fixed mouth embeddings of real and forged data in order to detect fake videos based on mouth movements without overfitting to low-level, manipulation-specific artefacts. Extensive experiments show that this simple approach significantly surpasses the state-of-the-art in terms of generalisation to unseen manipulations and robustness to perturbations, as well as shed light on the factors responsible for its performance." @default.
• W3110828520 created "2020-12-21" @default.
• W3110828520 creator A5009475700 @default.
• W3110828520 creator A5010477883 @default.
• W3110828520 creator A5016033078 @default.
• W3110828520 creator A5079503255 @default.
• W3110828520 date "2020-12-14" @default.
• W3110828520 modified "2023-09-27" @default.
• W3110828520 title "Lips Don't Lie: A Generalisable and Robust Approach to Face Forgery Detection" @default.
• W3110828520 cites W1522301498 @default.
• W3110828520 cites W1677182931 @default.
• W3110828520 cites W1849277567 @default.
• W3110828520 cites W2064675550 @default.
• W3110828520 cites W2095705004 @default.
• W3110828520 cites W2099471712 @default.
• W3110828520 cites W2194775991 @default.
• W3110828520 cites W2301937176 @default.
• W3110828520 cites W2412509443 @default.
• W3110828520 cites W2531409750 @default.
• W3110828520 cites W2556882396 @default.
• W3110828520 cites W2594690981 @default.
• W3110828520 cites W2619947201 @default.
• W3110828520 cites W2737804889 @default.
• W3110828520 cites W2777769595 @default.
• W3110828520 cites W2888339491 @default.
• W3110828520 cites W2891145043 @default.
• W3110828520 cites W2902304528 @default.
• W3110828520 cites W2903369540 @default.
• W3110828520 cites W2906447902 @default.
• W3110828520 cites W2911424785 @default.
• W3110828520 cites W2912336782 @default.
• W3110828520 cites W2914447220 @default.
• W3110828520 cites W2916798096 @default.
• W3110828520 cites W2942074357 @default.
• W3110828520 cites W2949117887 @default.
• W3110828520 cites W2950635152 @default.
• W3110828520 cites W2959650063 @default.
• W3110828520 cites W2963155035 @default.
• W3110828520 cites W2963420686 @default.
• W3110828520 cites W2963446712 @default.
• W3110828520 cites W2963773039 @default.
• W3110828520 cites W2964222566 @default.
• W3110828520 cites W2969985801 @default.
• W3110828520 cites W2972511899 @default.
• W3110828520 cites W2979894294 @default.
• W3110828520 cites W2979980060 @default.
• W3110828520 cites W2980037812 @default.
• W3110828520 cites W2981803355 @default.
• W3110828520 cites W2982058372 @default.
• W3110828520 cites W2984287396 @default.
• W3110828520 cites W2984700035 @default.
• W3110828520 cites W2996179709 @default.
• W3110828520 cites W2999135330 @default.
• W3110828520 cites W3012033723 @default.
• W3110828520 cites W3012472557 @default.
• W3110828520 cites W3016011581 @default.
• W3110828520 cites W3034196597 @default.
• W3110828520 cites W3034214108 @default.
• W3110828520 cites W3034301684 @default.
• W3110828520 cites W3034577585 @default.
• W3110828520 cites W3034713808 @default.
• W3110828520 cites W3034864980 @default.
• W3110828520 cites W3034900344 @default.
• W3110828520 cites W3035063907 @default.
• W3110828520 cites W3035446294 @default.
• W3110828520 cites W3046589207 @default.
• W3110828520 cites W3048341421 @default.
• W3110828520 cites W3083246145 @default.
• W3110828520 cites W3087162391 @default.
• W3110828520 cites W3090124563 @default.
• W3110828520 cites W3093010840 @default.
• W3110828520 cites W3093077034 @default.
• W3110828520 cites W3094728142 @default.
• W3110828520 cites W3095986900 @default.
• W3110828520 cites W3097792222 @default.
• W3110828520 cites W3104792420 @default.
• W3110828520 cites W3108281670 @default.
• W3110828520 cites W3113824005 @default.
• W3110828520 hasPublicationYear "2020" @default.
• W3110828520 type Work @default.
• W3110828520 sameAs 3110828520 @default.
• W3110828520 citedByCount "1" @default.
• W3110828520 countsByYear W31108285202021 @default.
• W3110828520 crossrefType "posted-content" @default.
• W3110828520 hasAuthorship W3110828520A5009475700 @default.
• W3110828520 hasAuthorship W3110828520A5010477883 @default.
• W3110828520 hasAuthorship W3110828520A5016033078 @default.
• W3110828520 hasAuthorship W3110828520A5079503255 @default.
• W3110828520 hasConcept C104317684 @default.
• W3110828520 hasConcept C108583219 @default.
• W3110828520 hasConcept C144024400 @default.
• W3110828520 hasConcept C154945302 @default.
• W3110828520 hasConcept C185592680 @default.
• W3110828520 hasConcept C22019652 @default.
• W3110828520 hasConcept C2779304628 @default.
• W3110828520 hasConcept C28490314 @default.
• W3110828520 hasConcept C31972630 @default.
• W3110828520 hasConcept C36289849 @default.
• W3110828520 hasConcept C41008148 @default.
• W3110828520 hasConcept C50644808 @default.

• next