A Lightweight and Effective Image Tampering Localization Network with Vision Mamba

TL;DR

This paper introduces ForMa, a lightweight image tampering localization network leveraging vision Mamba for efficient global feature extraction, achieving state-of-the-art results with low computational cost.

Contribution

The paper proposes a novel ForMa network based on vision Mamba, combining multi-scale global features and a noise-assisted decoder for improved tampering localization.

Findings

Achieves state-of-the-art generalization on 10 datasets.

Maintains lowest computational complexity among peers.

Demonstrates robustness and efficiency in tampering detection.

Abstract

Current image tampering localization methods primarily rely on Convolutional Neural Networks (CNNs) and Transformers. While CNNs suffer from limited local receptive fields, Transformers offer global context modeling at the expense of quadratic computational complexity. Recently, the state space model Mamba has emerged as a competitive alternative, enabling linear-complexity global dependency modeling. Inspired by it, we propose a lightweight and effective FORensic network based on vision MAmba (ForMa) for blind image tampering localization. Firstly, ForMa captures multi-scale global features that achieves efficient global dependency modeling through linear complexity. Then the pixel-wise localization map is generated by a lightweight decoder, which employs a parameter-free pixel shuffle layer for upsampling. Additionally, a noise-assisted decoding strategy is proposed to integrate complementary manipulation traces from tampered images, boosting decoder sensitivity to forgery cues. Experimental results on 10 standard datasets demonstrate that ForMa achieves state-of-the-art generalization ability and robustness, while maintaining the lowest computational complexity. Code is available at https://github.com/multimediaFor/ForMa.