TBFormer: Two-Branch Transformer for Image Forgery Localization
Authors:
- Yaqi LiuYLYaqi Liu
- Binbin LvBLBinbin Lv
- Xin JinXJXin Jin
- Xiaoyu ChenXCXiaoyu Chen
- Xiaokun ZhangXZXiaokun Zhang
Preprint
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.Keywords
All Related Versions
- Version 1, 2023-02-25, ArXiv
- Published version: IEEE Signal Processing Letters, 30 (10709908), 623.
This publication has 13 references indexed in Scilit:
- Image splicing forgery detection by combining synthetic adversarial networks and hybrid dense U‐net based on multiple spacesInternational Journal of Intelligent Systems, 2022
- ET: Edge-Enhanced Transformer for Image Splicing DetectionIEEE Signal Processing Letters, 2022
- Exposing splicing forgery in realistic scenes using deep fusion networkInformation Sciences, 2020
- An End-to-End Dense-InceptionNet for Image Copy-Move Forgery DetectionIEEE Transactions on Information Forensics and Security, 2019
- Adversarial Learning for Constrained Image Splicing Detection and Localization Based on Atrous ConvolutionIEEE Transactions on Information Forensics and Security, 2019
- Semantic Understanding of Scenes Through the ADE20K DatasetInternational Journal of Computer Vision, 2018
- A deep learning approach to patch-based image inpainting forensicsSignal Processing: Image Communication, 2018
- Constrained Convolutional Neural Networks: A New Approach Towards General Purpose Image Manipulation DetectionIEEE Transactions on Information Forensics and Security, 2018
- Multi-Scale Analysis Strategies in PRNU-Based Tampering LocalizationIEEE Transactions on Information Forensics and Security, 2016
- Efficient Dense-Field Copy–Move Forgery DetectionIEEE Transactions on Information Forensics and Security, 2015