Multi-domain Analysis and Generalization of Image manipulation loCalization
Keywords: Domain generalization, Diffusion-Based Inpainting, Misinformation Detection, Computer Vision, Benchmark Dataset, Visual Forensics
Abstract: Advanced image editing software enables easy creation of highly convincing image manipulations, which has been made even more accessible in recent years due to advances in generative AI. Manipulated images, while often harmless, could spread misinformation, create false narratives, and influence people’s opinions on important issues. Despite this growing threat, current research on detecting advanced manipulations across different visual domains, remains limited. Thus, we introduce Multi-domain Analysis and Generalization of Image manipulation loCalization (MAGIC), a comprehensive benchmark designed for studying generalization across several axes in image manipulation detection. MAGIC comprises over 192K images from two distinct sources (user and news photos), spanning a diverse range of topics and manipulation sizes. We focus on images manipulated using recent diffusion-based inpainting methods, which are largely absent in existing datasets. We conduct experiments under different types of domain shift to evaluate robustness of existing image manipulation detection methods. Our goal is to drive further research in this area by offering new insights that would help develop more reliable and generalizable image manipulation detection methods. We will release the dataset after this work is published.
Primary Area: datasets and benchmarks
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 7271
Loading