2026· IEEE Signal Processing Letters· Vol 33, pp. 2834-2838· 0 citations· 45 references
Computer Science
Abstract
In general, the cross-generator generalization and robustness against attacks are two main challenges in AI-generated image detection. To address this, we put forward a synthetic image detector (DTBF), a two-branch architecture followed by a jointly-optimized concatenation (JOC), aiming at capturing low-level artifacts and high-level semantics and dynamically fusing them to enhance the generalization and robustness. In the artifact-extracted branch (AE-branch), the local multi-direction binary-encoding patterns (LMBP) are customized to extract and encode the relationships among pixels within each sliding window, resulting in the LMBP distribution serving as a universal fingerprint to distinguish real/fake images. The dual-alignment guided semantic branch (DAS-branch), working as a supplement to AE-branch, designs global context-unrelated prompts and semantic-enhanced prompts to capture global semantic inconsistency and local patch anomalies, strengthening the generalization and robustness of DTBF. Finally, JOC dynamically fuses the two branches through concatenation to amplify the effectiveness of each branch, achieving better generalization and robustness. With the assistance of two complementary branches and JOC, our proposed DTBF significantly outperforms 12 state-of-the-art detectors on two publicly available datasets in terms of detection accuracy and robustness.
Diffusion-based inpainting models modify only a localized part of an image, while many AI-image detectors rely on global artifacts and do not localize. These artifacts vary across generators, limiting detector transfer under distribution shifts. Recent work shows that restoring the authentic pixels outside the inpainted region removes these cues and can degrade pretrained detectors. To address this, we present FUSED, a unified framework for the joint detection and localization of AI-generated inpainting. FUSED combines low-level forensic cues with high-level semantic features using a sparsely-gated Mixture-of-Experts architecture, enabling the model to adaptively prioritize the most relevant signal for each token. For each input, FUSED predicts both an image-level manipulation score and a pixel-level mask of the inpainted area. On the OpenSDID cross-generator benchmark, FUSED achieves the best average detection and localization, with the largest gains on unseen generators. The same model transfers directly to the held-out AutoSplice and CocoGlide benchmarks, more than doubling localization performance. Evaluating each held-out benchmark with and without the global generator artifact further shows that all evaluated methods, ours included, partly read the artifact as evidence of manipulation, and FUSED remains the strongest under both conditions. Code and pretrained models are available at https://github.com/AntonNuzhdin/FUSED.
Anton Nuzhdin, Marcel Worring, Ivona Najdenkoska· 0 citations
This work introduces a dual-branch ensemble framework fusing Semantic Deep Learning with Mathematical Forensic Feature Extraction, highlighting the practicality and scalability of mathematical forensics for real-world deployment.
AI-generated image (AIGI) detectors achieve strong accuracy on clean benchmarks, but their performance drops sharply after images are propagated through real-world channels. We trace this fragility to what these detectors actually learn: they overfit to local artifacts left by generators in small spatial neighborhoods, which are easily destroyed by common propagation degradations such as JPEG compression and blur. Instead, we shift the discriminative cue from fragile local artifacts to more robust global structure. Building on this, we propose GlobalForge, a framework with two complementary modules. The Local Information Bottleneck (LIB) suppresses local components to block shortcut learning, while the Global Structural Reasoning (GSR) module forces every token to gather evidence from distant regions. Both modules are trained jointly under a contrastive structural loss based on degradation that keeps the resulting features stable under degradation. To support fine-grained robustness evaluation, we further introduce RealDeg-Bench, covering 7 common degradation operators and multi-step compound chains. GlobalForge improves average BAcc on 8 in-the-wild benchmark groups by $\mathbf{5.89\%}$ over the previous state-of-the-art, and is clearly ahead of representative baselines on RealDeg-Bench under both single and compound degradations. Code is available at https://anonymous.4open.science/r/GlobalForge-BE0F/.
Manni Cui, Ruiqi Liu, Dianyuan Zou et al.· 0 citations
A data-generation pipeline that captions real photographs with a vision–language model and regenerates them with modern text-to-image systems, producing semantically aligned real/synthetic pairs that isolate generative artifacts from image content is described.
Deepfake detection remains a challenging problem due to the ever-evolving techniques used to synthesize realistic fake images and videos. Traditional detection methods primarily operate in the pixel domain, often struggling to generalize across different forgery types. In this work, we explore an approach that integrates wavelet-based frequency analysis with deep learning to enhance deepfake detection. By decomposing input images using the Discrete Wavelet Transform (DWT), we extract multi-resolution features that capture both global structures and fine-grained manipulation artifacts; such features are then input to a neural network. Our results show that level-1 DWT features mainly induce a sensitivity-specificity tradeoff: they improve fake-class sensitivity and F1 in cross-forgery settings, while RGB models retain stronger specificity and in-forgery performance. This suggests that wavelet sub-bands expose manipulation cues that are useful for detecting unseen fake classes, but they should not be interpreted as a uniform robustness improvement.
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.
Jiangling Zhang, Shuxuan Gao, Zeyu Chen et al.· 0 citations