EDGER: EDge-Guided with HEatmap Refinement for Generalizable Image Forgery Localization

TL;DR

EDGER is a novel framework for localizing manipulated regions in images, combining edge detection and heatmap refinement to improve accuracy and cross-domain generalization without losing image resolution.

Contribution

The paper introduces EDGER, a dual-branch, patch-based approach that fuses frequency-based edge detection with synthetic heatmapping for robust, resolution-agnostic forgery localization.

Findings

Effective in multi-megapixel images

Strong cross-domain generalization

Combines edge cues with synthetic priors

Abstract

Text-guided inpainting has made image forgery increasingly realistic, challenging both SID and IFL. However, existing methods often struggle to point out suspicious signals across domains. To address this problem, we propose EDGER, a patch-based, dual-branch framework that localizes manipulated regions in arbitrary resolution images without sacrificing native resolution. The first branch, Edge-Guided Segmentation, introduces a Frequency-based Edge Detector to emphasize high-frequency inconsistencies at manipulation boundaries, and fine-tunes a SegFormer to fuse RGB and edge features for pixel-level masks. Since edge evidence is most informative only when patches contain both authentic and manipulated pixels, we complement Edge-Guided Segmentation with a Synthetic Heatmapping branch, a classification-based localizer that fine-tunes a CLIP-ViT image encoder with LoRA to flag fully synthetic patches. Together, Synthetic Heatmapping provides coarse, patch-level synthetic priors, while Edge-Guided Segmentation sharpens boundaries within partially manipulated patches, yielding comprehensive localization. Evaluated in the MediaEval 2025, SynthIM challenge, Manipulated Region Localization Task's setting, our approach scales to multi-megapixel imagery and exhibits strong cross-domain generalization. Extensive ablations highlight the complementary roles of frequency-based edge cues and patch-level synthetic priors in driving accurate, resolution-agnostic localization.