FOCA: Frequency-Oriented Cross-Domain Forgery Detection, Localization and Explanation via Multi-Modal Large Language Model

TL;DR

FOCA is a multimodal framework that improves image forgery detection, localization, and explanation by integrating spatial and frequency domain features with a large language model, supported by a new dataset.

Contribution

The paper introduces FOCA, a novel multimodal large language model-based framework for forgery detection that combines spatial and frequency features with interpretability, along with a new dataset FSE-Set.

Findings

FOCA outperforms existing methods in detection accuracy.

FOCA provides human-interpretable explanations of forgeries.

Extensive experiments validate FOCA's effectiveness across domains.

Abstract

Advances in image tampering techniques, particularly generative models, pose significant challenges to media verification, digital forensics, and public trust. Existing image forgery detection and localization (IFDL) methods suffer from two key limitations: over-reliance on semantic content while neglecting textural cues, and limited interpretability of subtle low-level tampering traces. To address these issues, we propose FOCA, a multimodal large language model-based framework that integrates discriminative features from both the RGB spatial and frequency domains via a cross-attention fusion module. This design enables accurate forgery detection and localization while providing explicit, human-interpretable cross-domain explanations. We further introduce FSE-Set, a large-scale dataset with diverse authentic and tampered images, pixel-level masks, and dual-domain annotations. Extensive experiments show that FOCA outperforms state-of-the-art methods in detection performance and interpretability across both spatial and frequency domains.