EVAS localizes sparse multimodal forgeries via multi-stage audio-visual synergy and decoupled boundary-aware refinement, reporting SOTA AP and AR on LAV-DF, AV-Deepfake1M, and TVIL.
Title resolution pending
1 Pith paper cite this work, alongside 36 external citations. Polarity classification is still indexing.
1
Pith paper citing it
36
external citations · external index
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
EVAS: Efficient Multimodal Temporal Forgery Localization via Audio-Visual Synergy and Steered Boundary Calibration
EVAS localizes sparse multimodal forgeries via multi-stage audio-visual synergy and decoupled boundary-aware refinement, reporting SOTA AP and AR on LAV-DF, AV-Deepfake1M, and TVIL.