cs.CVAug 10, 2025

Bridging Semantic Logic Gaps: A Cognition Inspired Multimodal Boundary Preserving Network for Image Manipulation Localization

Authors: Songlin Li, Zhiqing Guo, Yuanman Li, Zeyu Li, Yunfeng Diao, Gaobo Yang, Liejun Wang

Organizations: the School of Computer Science and Technology, Xinjiang University, Urumqi 830046, China · the School of Electronic and Information Engineering, Shenzhen University. Shenzhen, China · the School of Computer Science and Information Engineering, Hefei University of Technology. Hefei, China · the College of Computer Science and Electronic Engineering, Hunan University, Changsha 410082, China

Abstract

The existing image manipulation localization (IML) models mainly relies on visual cues, but ignores the semantic logical relationships between content features. In fact, the content semantics conveyed by real images often conform to human cognitive laws. However, image manipulation technology usually destroys the internal relationship between content features, thus leaving semantic clues for IML. In this paper, we propose a cognition inspired multimodal boundary preserving network (CMB-Net). Specifically, CMB-Net utilizes large language models (LLMs) to analyze manipulated regions within images and generate prompt-based textual information to compensate for the lack of semantic relationships in the visual information. Considering that the erroneous texts induced by hallucination from LLMs will damage the accuracy of IML, we propose an image-text central ambiguity module (ITCAM). It assigns weights to the text features by quantifying the ambiguity between text and image features, thereby ensuring the beneficial impact of textual information. We also propose an image-text interaction module (ITIM) that aligns visual and text features using a correlation matrix for fine-grained interaction. Finally, inspired by invertible neural networks, we propose a restoration edge decoder (RED) that mutually generates input and output features to preserve boundary information in manipulated regions without loss. Extensive experiments show that CMB-Net outperforms most existing IML models. Our code is available on https://github.com/vpsg-research/CMB-Net.

Figures & tables

Explore similar work

CardsList
  1. Off-the-shelf Vision Models Benefit Image Manipulation Localization

    Apr 10, 2026Zhengxuan Zhang, Keji Song, Junmin Hu +2Image Forgery LocalizationVision Foundation Model Adaptation

  2. Can We Model the Artifacts Explicitly? Disentangle Artifacts via Pairwise Edit Relations for Image Manipulation Localization

    Oct 6, 2026Xuekang Zhu, Kaiwen Feng, Ruifeng Wang +10Image Forgery Localization

  3. Semantic Manipulation Localization

    Apr 11, 2026Zhenshan Tan, Chenhan Lu, Yuxiang Huang +6Image Forgery LocalizationDigital Image Forensics