FerretNet: Efficient Synthetic Image Detection via Local Pixel Dependencies

2025-09-25

AI Summary

The increasing photorealism of generative images poses significant challenges for reliable detection. To address this, we propose FerretNet—a lightweight detection network (1.1M parameters) that models local pixel dependencies, and the first to incorporate Markov Random Field (MRF) theory into synthetic image detection. FerretNet exploits inherent distributional biases in the generative process and decoding-induced smoothing effects to capture local inconsistencies in texture and edge structures. Its architecture integrates local dependency modeling, parameter-efficient convolutional design, and reconstruction error analysis. Trained exclusively on ProGAN-generated images from four classes, FerretNet achieves a mean accuracy of 97.1% on an open-world benchmark spanning 22 diverse generative models—surpassing state-of-the-art methods by 10.6 percentage points. This demonstrates both strong generalization across architectures and superior discriminative capability for subtle, localized artifacts.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

The increasing realism of synthetic images generated by advanced models such as VAEs, GANs, and LDMs poses significant challenges for synthetic image detection. To address this issue, we explore two artifact types introduced during the generation process: (1) latent distribution deviations and (2) decoding-induced smoothing effects, which manifest as inconsistencies in local textures, edges, and color transitions. Leveraging local pixel dependencies (LPD) properties rooted in Markov Random Fields, we reconstruct synthetic images using neighboring pixel information to expose disruptions in texture continuity and edge coherence. Building upon LPD, we propose FerretNet, a lightweight neural network with only 1.1M parameters that delivers efficient and robust synthetic image detection. Extensive experiments demonstrate that FerretNet, trained exclusively on the 4-class ProGAN dataset, achieves an average accuracy of 97.1% on an open-world benchmark comprising across 22 generative models, surpassing state-of-the-art methods by 10.6%.

Problem

• Detecting increasingly realistic synthetic images from advanced models
• Addressing latent distribution deviations and decoding-induced smoothing artifacts
• Exposing texture and edge inconsistencies via local pixel dependencies

Innovation

• Uses local pixel dependencies for detection
• Reconstructs images to expose texture disruptions
• Lightweight network with 1.1M parameters