SUMI-IFL: An Information-Theoretic Framework for Image Forgery Localization with Sufficiency and Minimality Constraints

TL;DR

SUMI-IFL introduces an information-theoretic framework for image forgery localization that ensures comprehensive and minimal feature representations, significantly improving accuracy and robustness against tampering.

Contribution

It proposes a novel framework imposing sufficiency and minimality constraints on forgery features, advancing the accuracy and robustness of image forgery localization methods.

Findings

Outperforms existing methods on in-dataset evaluations.

Achieves superior cross-dataset generalization.

Effectively captures comprehensive forgery clues.

Abstract

Image forgery localization (IFL) is a crucial technique for preventing tampered image misuse and protecting social safety. However, due to the rapid development of image tampering technologies, extracting more comprehensive and accurate forgery clues remains an urgent challenge. To address these challenges, we introduce a novel information-theoretic IFL framework named SUMI-IFL that imposes sufficiency-view and minimality-view constraints on forgery feature representation. First, grounded in the theoretical analysis of mutual information, the sufficiency-view constraint is enforced on the feature extraction network to ensure that the latent forgery feature contains comprehensive forgery clues. Considering that forgery clues obtained from a single aspect alone may be incomplete, we construct the latent forgery feature by integrating several individual forgery features from multiple perspectives. Second, based on the information bottleneck, the minimality-view constraint is imposed on the feature reasoning network to achieve an accurate and concise forgery feature representation that counters the interference of task-unrelated features. Extensive experiments show the superior performance of SUMI-IFL to existing state-of-the-art methods, not only on in-dataset comparisons but also on cross-dataset comparisons.