Multi-axis Analysis of Image Manipulation Localization

2026-05-19

AI Summary

This work addresses the limited generalization of existing image manipulation detection methods when confronted with generative AI–driven manipulations that span diverse domains, qualities, types, and scales, as well as the absence of a systematic evaluation benchmark. To bridge this gap, we introduce AUDITS—the first large-scale, structured benchmark for comprehensive image manipulation detection—integrating four key analytical dimensions: domain shift, image quality, manipulation type, and manipulation scale. Built upon 530,000 real-world user and news images, AUDITS leverages diffusion models to generate diverse inpainting-based forgeries and establishes a cross-domain, multi-scale evaluation paradigm. Extensive experiments reveal critical performance bottlenecks of current methods across these dimensions, offering essential data resources and insights to advance robust and generalizable manipulation localization techniques.

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, there is limited research on detecting advanced manipulations across different visual domains. Thus, we introduce Analysis Under Domain-shifts, qualIty, Type, and Size (AUDITS), a comprehensive benchmark designed for studying axes of analysis in image manipulation detection. AUDITS comprises over 530K images from two distinct sources (user and news photos). We curate our dataset to support analysis across multiple axes using recent diffusion-based inpaintings, spanning a diverse range of manipulation types and sizes. 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.

Problem

• image manipulation detection
• domain shift
• generative AI
• misinformation
• manipulation localization

Innovation

• image manipulation localization
• domain shift
• diffusion-based inpainting
• benchmark dataset
• multi-axis analysis