[PDF] PolyGlotFake: A Novel Multilingual and Multimodal DeepFake Dataset | Semantic Scholar (2024)

[PDF] Semantic Reader

Although Deep fake detection is extremely difficult and still an unsolved problem, a Deepfake detection model trained only on the DFDC can generalize to real "in-the-wild" Deepfake videos, and such a model can be a valuable analysis tool when analyzing potentially Deepfaked videos.

It is assertion that this database will improve the state-of-the-art by extending the capabilities of deepfake detection algorithms to real-world scenarios by proposing the DF-Platter dataset, which contains both low-resolution and high-resolution deepfakes generated using multiple generation techniques.

This work proposes a statistical consistency attack (StatAttack) against DeepFake detectors, and finds that the statistical differences between natural and DeepFake images are positively associated with the distribution shifting between the two kinds of images, and is proposed to use a distribution-aware loss to guide the optimization of different degradations.

Vall-E significantly outperforms the state-of-the-art zero-shot TTS system in terms of speech naturalness and speaker similarity, and is found to preserve the speaker's emotion and acoustic environment of the acoustic prompt in synthesis.

This paper adopts the end-to-end framework of VITS for high-quality waveform reconstruction, and proposes strategies for clean content information extraction without text annotation and the spectrogram-resize based data augmentation to improve the purity of extracted content information.

This paper proposes a forgery detection frame-work emphasizing the common compact representations of genuine faces based on reconstruction-classification learning, and builds bipartite graphs over the encoder and decoder features in a multi-scale fashion.

This work constrain the consistency of different representations explicitly and proposes a simple yet effective framework, COnsistent REpresentation Learning (CORE), which first captures the different representations with different augmentations, then regularize the cosine distance of the representations to enhance the consistency.

This paper proposes a frequency adversarial attack method against face forgery detectors that fools not only the spatial-based detectors but also the state-of- the-art frequency-based detector effectively and enhances the transferability across face forgeries detectors as black-box attacks.

...

...