aiqbench analysis
This research addresses the critical challenge of detecting GAN-generated deepfakes by optimizing detection architectures for efficiency without sacrificing accuracy. By hybridizing EfficientNet's convolutional strengths with Swin Transformer attention mechanisms, the study achieves near-perfect recall while reducing computational overhead compared to standard Vision Transformers. This approach offers a practical pathway for deploying lightweight AI models in real-time forensic applications where resource constraints are significant.
Swin Meets EfficientNet: Lightweight Architectures for GAN-Based Face Forensics
arXiv:2609.01749v1 Announce Type: cross Abstract: 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.
Abstract: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.
12 pages, 2 figures, 1 table. Presented at the International Conference on Computational Techniques in Data Science (IICTDS 2025), online. Proceedings forthcoming
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
arXiv:2609.01749 [cs.CV]
(or arXiv:2609.01749v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2609.01749 arXiv-issued DOI via DataCite (pending registration)
From: Sahil Sharma Dr. [view email] [v1] Tue, 1 Sep 2026 18:19:32 UTC (5,185 KB)
Full text under the publisher's open licence · 289 words
Related coverage
A Red Line and Oversight Framework for Government AI Contracts
AI Alignment Forum · Jul 18, 2026
.webp%3Fssl%3D1&w=320)
Top 50 Best Penetration Testing Companies in 2026
Cybersecurity News · Jul 31, 2026
How to Build In-Vehicle AI Agents with NVIDIA: From Cloud to Car
NVIDIA · May 5, 2026
Article: Architecting Secure and Scalable Facial Verification Systems
InfoQ · Sep 18, 2026