# TBFormer: Two-Branch Transformer for Image Forgery Localization

## TL;DR

TBFormer is a novel two-branch Transformer network that effectively localizes image forgeries by combining RGB and noise domain features through hierarchical fusion and attention mechanisms, achieving superior results.

## Contribution

Introduces a two-branch Transformer architecture with hierarchical feature fusion and attention for improved image forgery localization.

## Key findings

- Effective in capturing subtle forgery traces
- Outperforms existing methods on benchmark datasets
- Demonstrates robustness across different forgery types

## Abstract

Image forgery localization aims to identify forged regions by capturing subtle traces from high-quality discriminative features. In this paper, we propose a Transformer-style network with two feature extraction branches for image forgery localization, and it is named as Two-Branch Transformer (TBFormer). Firstly, two feature extraction branches are elaborately designed, taking advantage of the discriminative stacked Transformer layers, for both RGB and noise domain features. Secondly, an Attention-aware Hierarchical-feature Fusion Module (AHFM) is proposed to effectively fuse hierarchical features from two different domains. Although the two feature extraction branches have the same architecture, their features have significant differences since they are extracted from different domains. We adopt position attention to embed them into a unified feature domain for hierarchical feature investigation. Finally, a Transformer decoder is constructed for feature reconstruction to generate the predicted mask. Extensive experiments on publicly available datasets demonstrate the effectiveness of the proposed model.