TL;DR
This study investigates the impact of image reconstruction processes using two approaches: Generative Adversarial Networks (GANs) and Variational Autoencoders (VAEs) on authentic and forged images, and establishes a foundational understanding of forensic artifacts resulting from the reconstruction process.
Digital image forgery has become a critical concern in the era of advanced multimedia technologies, where the authenticity of visual content directly affects trust in digital communication, journalism, and law enforcement. Among various forgery techniques, copy-move forgery (CMF) is among the most common and deceptive, as it involves duplicating a region of an image to conceal or misrepresent information. To address this challenge, numerous copy-move forgery detection (CMFD) approaches have been proposed, ranging from block-based and keypoint-based methods to hybrid models and deep learning (DL) techniques. This paper provides a comprehensive review of these approaches, analyzing their strengths and limitations, and evaluating their performance across multiple benchmark datasets. The evaluation considers factors such as image resolution, manipulation types, and robustness against post-processing attacks. By systematically comparing the algorithms and datasets, the study highlights persistent challenges and outlines future research directions. The findings aim to guide researchers in selecting appropriate techniques and inspire the development of more robust CMFD solutions.
Overall, GenPix provides a challenging and realistic benchmark for evaluating modern detectors, and the proposed AAE offers an efficient, interpretable baseline for future research on general-purpose fake-image detection.
A generative detector that localizes tampering by estimating the local restoration cost required to align a query image with authentic visual-text statistics, rather than by learning forgery-specific decision boundaries is proposed, and Sparse-Constraint Rectified Flow is introduced, a detector-oriented adaptation of Flow Matching for spatially sparse anomaly localization.
The rapid advancement of diffusion models has raised concerns about their misuse in generating deceptive visual content, motivating growing interest in AI-generated image detection. Many existing detection methods rely on image semantic features, but modern diffusion models are optimized to closely match the semantic structure of real images, reducing the effectiveness of semantic-based detection. An alternative line of work exploits differences revealed through diffusion reconstruction; however, most existing approaches treat reconstruction error as a static and passive metric, which can be sensitive to generators or post-processing, thereby limiting robustness. In this work, we propose Adversarial Diffusion Reconstruction Distance (ADRD), a detection framework that models diffusion reconstruction as a dynamic response process rather than a fixed descriptor. ADRD actively probes the reconstruction behavior by introducing perturbation in latent space and measuring how reconstruction deviations respond under identical perturbations. We empirically observe that real images typically exhibit larger and more variable reconstruction responses, while diffusion-generated images tend to show more stable reconstruction behavior, reflecting differences in their alignment with the diffusion model’s implicit data manifold. By characterizing reconstruction sensitivity instead of absolute reconstruction error, ADRD provides a complementary perspective to existing reconstruction-based detectors. Experimental evaluations on multiple benchmarks suggest that reconstruction response under controlled perturbations constitutes a meaningful signal for diffusion-generated image detection. The code is available at https://github.com/ezell-chou/adrd
This work proposes two novel attack methods targeted at detectors that leverage autoencoder reconstruction error and finds that by constructing imperceptible adversarial examples, the distance between original and reconstruction can be artificially increased, causing fake images to be wrongly classified as real.
We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.
