EUVIMEDEUVIMED entdecken
Crossref

Deep Network Representations as Reliable Indicators of Synthetic Content in Audiovisual and Clinical Contexts

Karol JĘDRASIAK, Julia BIJOCH

IBIMA Business Review · 01.04.2026

DOI: 10.5171/2025.4638225 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IBIMA Business Review, online erschienen am 01.04.2026. Verfasst von Karol JĘDRASIAK und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study introduces an interpretable framework for detecting synthetic audiovisual content using deep neural representations, applied to the DeepFake RealWorld (DFRW) dataset (46 371 clips; 77% with audio). Visual, acoustic, and cross-modal embeddings from ResNet, Vision Transformer, SlowFast, Wav2Vec2, and ECAPA-TDNN were evaluated with frequency-based metrics (Δp ≥ 0.15, PR ≥ 1.5). The strongest indicators were facial embedding variance (Δp = 0.29, PR = 3.4), Mahalanobis distance (Δp = 0.25), and audiovisual coherence (Δp = 0.23), all stable within 15% under compression and re-capture. In teledentistry and telemedicine, such explainable AI markers enhance authenticity verification of digital evidence and strengthen medico-legal reliability.

Zitieren

APA

JĘDRASIAK, K., & BIJOCH, J. (2026). Deep Network Representations as Reliable Indicators of Synthetic Content in Audiovisual and Clinical Contexts. IBIMA Business Review. https://doi.org/10.5171/2025.4638225

Vancouver

JĘDRASIAK K, BIJOCH J. Deep Network Representations as Reliable Indicators of Synthetic Content in Audiovisual and Clinical Contexts. IBIMA Business Review. 2026. doi:10.5171/2025.4638225

BibTeX
@article{JDRASIAK2026,
  author = {JĘDRASIAK, Karol and BIJOCH, Julia},
  title = {Deep Network Representations as Reliable Indicators of Synthetic Content in Audiovisual and Clinical Contexts},
  journal = {IBIMA Business Review},
  year = {2026},
  doi = {10.5171/2025.4638225}
}
Kostenlos starten →