DcDsDiff: Dual-Conditional and Dual-Stream Diffusion Model for Generative Image Tampering Localization

Hao, Qixian; Niu, Shaozhang; Zhang, Jiwei; Wang, Kai

doi:10.24963/ijcai.2025/120

DcDsDiff: Dual-Conditional and Dual-Stream Diffusion Model for Generative Image Tampering Localization

Qixian Hao, Shaozhang Niu, Jiwei Zhang, Kai Wang

Download BibTeX Cite

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence. Main Track. Pages 1071-1079.

https://doi.org/10.24963/ijcai.2025/120

Keywords

Computer Vision: CV: Recognition (object detection, categorization)
Computer Vision: CV: Multimodal learning
Computer Vision: CV: Segmentation, grouping and shape analysis

Abstract

Generative Image Tampering (GIT), due to its high diversity and realism, poses a significant challenge to traditional image tampering localization techniques. Consequently, this paper introduces a denoising diffusion probabilistic model-based DcDsDiff, which comprises a Dual-View Conditional Network (DVCN) and a Dual-Stream Denoising Network (DSDN). DVCN provides clues about the tampered areas. It extracts tampering features in the high-frequency view and integrates them with spatial domain features using attention mechanisms. DSDN jointly generates mask image and detail image, enhancing the generalization capability of the model against new tampering forms through iterative denoising. A multi-stream interaction mechanism enables the two generative tasks to promote each other, prompting the model to generate localization results that are rich in detail and complete. Experiments show that DcDsDiff outperforms mainstream methods in accurate localization, generalization, extensibility, and robustness. Code page: https://github.com/QixianHao/DcDsDiff-and-GIT10K.

Qixian Hao, Shaozhang Niu, Jiwei Zhang, and Kai Wang. 2025. DcDsDiff: Dual-Conditional and Dual-Stream Diffusion Model for Generative Image Tampering Localization. In Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25. IJCAI, 1071–1079. https://doi.org/10.24963/ijcai.2025/120

@inproceedings{ijcai2025p120,
  title     = {DcDsDiff: Dual-Conditional and Dual-Stream Diffusion Model for Generative Image Tampering Localization},
  author    = {Hao, Qixian and Niu, Shaozhang and Zhang, Jiwei and Wang, Kai},
  booktitle = {Proceedings of the Thirty-Fourth International Joint Conference on
               Artificial Intelligence, {IJCAI-25}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {James Kwok},
  pages     = {1071--1079},
  year      = {2025},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2025/120},
  url       = {https://doi.org/10.24963/ijcai.2025/120},
}

TY  - CPAPER
TI  - DcDsDiff: Dual-Conditional and Dual-Stream Diffusion Model for Generative Image Tampering Localization
AU  - Hao, Qixian
AU  - Niu, Shaozhang
AU  - Zhang, Jiwei
AU  - Wang, Kai
T2  - Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence, IJCAI-25
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2025
SP  - 1071
EP  - 1079
DO  - 10.24963/ijcai.2025/120
UR  - https://doi.org/10.24963/ijcai.2025/120
ER  -