{"slug": "swin-meets-efficientnet-lightweight-architectures-for-gan-based-face-forensics", "title": "Swin Meets EfficientNet: Lightweight Architectures for GAN-Based Face Forensics", "summary": "Researchers proposed lightweight Swin-Transformer-based architectures for detecting GAN-generated synthetic faces, with a hybrid EfficientNet-B0+Swin model achieving 99% accuracy and 99.44% recall on 5,000 test images from the 140K Real and Fake Faces dataset, outperforming pure Swin variants and a CNN-only baseline. The study, posted on arXiv (2609.01749v1), addresses the growing difficulty of identifying forged facial images produced by generative models.", "body_md": "arXiv:2609.01749v1 Announce Type: new\nAbstract: 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.", "url": "https://wpnews.pro/news/swin-meets-efficientnet-lightweight-architectures-for-gan-based-face-forensics", "canonical_source": "https://arxiv.org/abs/2609.01749", "published_at": "2026-09-03 04:00:00+00:00", "updated_at": "2026-09-03 04:22:47.613854+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "computer-vision", "generative-ai"], "entities": ["arXiv", "Swin Transformer", "EfficientNet-B0", "StyleGAN", "Flickr", "DFDC", "140K Real and Fake Faces dataset"], "alternates": {"html": "https://wpnews.pro/news/swin-meets-efficientnet-lightweight-architectures-for-gan-based-face-forensics", "markdown": "https://wpnews.pro/news/swin-meets-efficientnet-lightweight-architectures-for-gan-based-face-forensics.md", "text": "https://wpnews.pro/news/swin-meets-efficientnet-lightweight-architectures-for-gan-based-face-forensics.txt", "jsonld": "https://wpnews.pro/news/swin-meets-efficientnet-lightweight-architectures-for-gan-based-face-forensics.jsonld"}}