CBDiff:Conditional Bernoulli Diffusion Models for Image Forgery Localization

TL;DR

CBDiff introduces a probabilistic diffusion model that generates multiple plausible forgery localization maps, improving reliability and capturing uncertainty in image tampering detection.

Contribution

It presents a novel Conditional Bernoulli Diffusion Model with Bernoulli noise and Time-Step Cross-Attention for enhanced image forgery localization.

Findings

Outperforms state-of-the-art methods on eight benchmark datasets

Produces diverse and plausible localization maps

Effectively models uncertainty in tampered regions

Abstract

Image Forgery Localization (IFL) is a crucial task in image forensics, aimed at accurately identifying manipulated or tampered regions within an image at the pixel level. Existing methods typically generate a single deterministic localization map, which often lacks the precision and reliability required for high-stakes applications such as forensic analysis and security surveillance. To enhance the credibility of predictions and mitigate the risk of errors, we introduce an advanced Conditional Bernoulli Diffusion Model (CBDiff). Given a forged image, CBDiff generates multiple diverse and plausible localization maps, thereby offering a richer and more comprehensive representation of the forgery distribution. This approach addresses the uncertainty and variability inherent in tampered regions. Furthermore, CBDiff innovatively incorporates Bernoulli noise into the diffusion process to more faithfully reflect the inherent binary and sparse properties of forgery masks. Additionally, CBDiff introduces a Time-Step Cross-Attention (TSCAttention), which is specifically designed to leverage semantic feature guidance with temporal steps to improve manipulation detection. Extensive experiments on eight publicly benchmark datasets demonstrate that CBDiff significantly outperforms existing state-of-the-art methods, highlighting its strong potential for real-world deployment.