Pixel-Inconsistency Modeling for Image Manipulation Localization
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2025-06 · 47 (6) · 4455-4472
DOI: 10.1109/tpami.2025.3541028 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025-06. Verfasst von Chenqi Kong und 5 weiteren Autor:innen. Die Arbeit wurde bisher 79-mal zitiert (Crossref) und nennt 102 Literaturangaben. Gefördert von: Rapid-Rich Object Search (ROSE) Lab, Nanyang Technological University, Agency for Science, Technology and Research.
Abstract (Originalsprache)
Digital image forensics plays a crucial role in image authentication and manipulation localization. Despite the progress powered by deep neural networks, existing forgery localization methodologies exhibit limitations when deployed to unseen datasets and perturbed images (i.e., lack of generalization and robustness to real-world applications). To circumvent these problems and aid image integrity, this paper presents a generalized and robust manipulation localization model through the analysis of pixel inconsistency artifacts. The rationale is grounded on the observation that most image signal processors (ISP) involve the demosaicing process, which introduces pixel correlations in pristine images. Moreover, manipulating operations, including splicing, copy-move, and inpainting, directly affect such pixel regularity. We, therefore, first split the input image into several blocks and design masked self-attention mechanisms to model the global pixel dependency in input images. Simultaneously, we optimize another local pixel dependency stream to mine local manipulation clues within input forgery images. In addition, we design novel Learning-to-Weight Modules (LWM) to combine features from the two streams, thereby enhancing the final forgery localization performance. To improve the training process, we propose a novel Pixel-Inconsistency Data Augmentation (PIDA) strategy, driving the model to focus on capturing inherent pixel-level artifacts instead of mining semantic forgery traces. This work establishes a comprehensive benchmark integrating 16 representative detection models across 12 datasets. Extensive experiments show that our method successfully extracts inherent pixel-inconsistency forgery fingerprints and achieve state-of-the-art generalization and robustness performances in image manipulation localization.
Förderung
- Rapid-Rich Object Search (ROSE) Lab
- Nanyang Technological University
- Agency for Science, Technology and Research
- OTS Research Programme
- RGC General Research Fund
- Research Grant Council of Hong Kong
- FAPESP Horus
Zitieren
APA
Kong, C., Luo, A., Wang, S., Li, H., Rocha, A., & Kot, A. C. (2025). Pixel-Inconsistency Modeling for Image Manipulation Localization. IEEE Transactions on Pattern Analysis and Machine Intelligence, 47(6), 4455-4472. https://doi.org/10.1109/tpami.2025.3541028
Vancouver
Kong C, Luo A, Wang S, Li H, Rocha A, Kot AC. Pixel-Inconsistency Modeling for Image Manipulation Localization. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2025;47(6):4455-4472. doi:10.1109/tpami.2025.3541028
BibTeX
@article{Kong2025,
author = {Kong, Chenqi and Luo, Anwei and Wang, Shiqi and Li, Haoliang and Rocha, Anderson and Kot, Alex C},
title = {Pixel-Inconsistency Modeling for Image Manipulation Localization},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2025},
volume = {47},
number = {6},
pages = {4455--4472},
doi = {10.1109/tpami.2025.3541028}
}
EUVIMED