Towards More Accurate Fake Detection on Images Generated from Advanced Generative and Neural Rendering Models

TL;DR

This paper introduces a novel unsupervised spectral-based detection method for identifying images generated by advanced neural rendering techniques, addressing the challenge of detecting high-fidelity synthetic images with improved generalization.

Contribution

It proposes a spectral domain feature extraction approach combined with spatial information for robust fake image detection, and creates a new diverse neural rendering image database for evaluation.

Findings

Superior detection accuracy on neural rendering generated images

Enhanced generalization to unseen synthetic image types

Effective spectral-spatial feature integration

Abstract

The remarkable progress in neural-network-driven visual data generation, especially with neural rendering techniques like Neural Radiance Fields and 3D Gaussian splatting, offers a powerful alternative to GANs and diffusion models. These methods can produce high-fidelity images and lifelike avatars, highlighting the need for robust detection methods. In response, an unsupervised training technique is proposed that enables the model to extract comprehensive features from the Fourier spectrum magnitude, thereby overcoming the challenges of reconstructing the spectrum due to its centrosymmetric properties. By leveraging the spectral domain and dynamically combining it with spatial domain information, we create a robust multimodal detector that demonstrates superior generalization capabilities in identifying challenging synthetic images generated by the latest image synthesis techniques. To address the absence of a 3D neural rendering-based fake image database, we develop a comprehensive database that includes images generated by diverse neural rendering techniques, providing a robust foundation for evaluating and advancing detection methods.