Extending MesoNet to six convolutional layers and swapping the deepfake training images for Stable Diffusion images raises three-class accuracy from 43% to 76% on a small in-distribution dataset, without external validation.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Detecting Facial Image Manipulations with Multi-Layer CNN Models
Extending MesoNet to six convolutional layers and swapping the deepfake training images for Stable Diffusion images raises three-class accuracy from 43% to 76% on a small in-distribution dataset, without external validation.