Uhlenbrock L, Cozzolino D, Moussa D, Verdoliva L, Riess C (2024)
Publication Language: English
Publication Type: Conference contribution, Conference Contribution
Publication year: 2024
Publisher: Association for Computing Machinery, Inc
Pages Range: 47-52
Conference Proceedings Title: Proceedings of the 2024 ACM Workshop on Information Hiding and Multimedia Security
ISBN: 9798400706370
Open Access Link: https://faui1-files.cs.fau.de/public/publications/mmsec/2024-Uhlenbrock-IHMMSec.pdf
High-quality artificially generated images are widely available now and increasingly realistic, posing challenges for image forensics in distinguishing them from real ones. Unfortunately, building a single detector that generalizes well to unseen generators is very difficult, creating the need for diverse cues. In this paper, we show that natural and synthetic images differ in their color statistics, possibly due to the widely used perceptual loss, which is more sensitive to brightness than to chroma differences. Consequently, color statistics offer valuable cues for forensic analysis and the development of robust detectors. Our experiments using simple hand-crafted color functions with a random forest achieve 91% accuracy averaged over all tested Diffusion Models, even with limited training samples.
APA:
Uhlenbrock, L., Cozzolino, D., Moussa, D., Verdoliva, L., & Riess, C. (2024). Did You Note My Palette? Unveiling Synthetic Images Through Color Statistics. In Association for Computing Machinery (Eds.), Proceedings of the 2024 ACM Workshop on Information Hiding and Multimedia Security (pp. 47-52). Baiona, ES: Association for Computing Machinery, Inc.
MLA:
Uhlenbrock, Lea, et al. "Did You Note My Palette? Unveiling Synthetic Images Through Color Statistics." Proceedings of the ACM Workshop on Information Hiding and Multimedia Security, Baiona Ed. Association for Computing Machinery, Association for Computing Machinery, Inc, 2024. 47-52.
BibTeX: Download