Metraipolis › archive › dc_d2f39cc95ad5

Swin Meets EfficientNet: Lightweight Architectures for GAN-Based Face Forensics — arXiv:2609.01749

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01749
License (as archived)
CC0-1.0
Archived
2026-10-17T03:26:49.255Z
SHA-256 of the archived text
7574dbeff2d04a7e730eb7c34d16011b87a188de7224545aac7bc6a3067397cf
Stable URL of this page
https://metraipolis.com/documento/dc_d2f39cc95ad5

How to cite

Swin Meets EfficientNet: Lightweight Architectures for GAN-Based Face Forensics — arXiv:2609.01749 — arXiv (en), https://arxiv.org/abs/2609.01749 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d2f39cc95ad5 (Metraipolis archive dc_d2f39cc95ad5, sha256 7574dbeff2d04a7e730eb7c34d16011b87a188de7224545aac7bc6a3067397cf)
BibTeX
@misc{dcd2f39cc95ad5,
  title = {Swin Meets EfficientNet: Lightweight Architectures for GAN-Based Face Forensics — arXiv:2609.01749},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01749},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d2f39cc95ad5; sha256 7574dbeff2d04a7e730eb7c34d16011b87a188de7224545aac7bc6a3067397cf},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01749. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Modern generative models, such as GANs, diffusion architectures, and autoregressive systems, now produce facial images that are nearly indistinguishable from authentic photographs. This capability makes detecting forged images increasingly difficult, raising serious concerns about identity theft, fraud, and misinformation campaigns. Our research focuses specifically on GAN-generated synthetic faces, which underpin many face-centric deepfakes, and investigates efficient detection approaches using image analysis alone. Existing detection systems rely heavily on either convolutional neural networks (CNNs) or global vision transformers. While CNNs excel at identifying texture-based local features, they struggle with broader contextual understanding. Traditional Vision Transformer (ViT) models can capture long-range structures effectively, but demand substantial computational resources. Our work explores Swin-Transformer-based architectures across three implementations: a compact Swin Transformer trained from the ground up, ImageNet-1K pre-trained Swin-Tiny and Swin-Small models adapted for binary classification, and a novel hybrid combining EfficientNet-B0's convolutional processing with a Swin Transformer backend. We evaluated all models using the 140K Real and Fake Faces dataset, which includes StyleGAN-generated fake faces alongside authentic images from Flickr and DFDC, with balanced splits for training, validation, and testing. The EfficientNetB0+Swin hybrid achieved 99% accuracy and a 99.44% recall on 5,000 test images, outperforming both pure Swin variants and a previous CNN-only baseline on this dataset. Our results suggest that combining hierarchical CNN features with shifted-window self-attention provides an efficient and computationally lightweight method for detecting GAN-generated synthetic faces.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-03 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.01749 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T03:13:12.477Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-03?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).