See ./requirements.txt And refer to official MMSegmentation documentation.

Prepare for the training dataset like VOCdevkit and move it into ./data/. The pre-trained model for backbone can be downloaded as following: Google Drive Link or Baiduyun Link(extract code: 0yh3).

python train.py

Download the weights from Google Drive Link or Baiduyun Link (extract code: jjyl) and move it into the ./checkpoints/. To run all images in "samples/" directory, run:

python test.py

@article{zhu2024effective,
 title={Effective image tampering localization with multi-scale convnext feature fusion},
 author={Zhu, Haochen and Cao, Gang and Zhao, Mo and Tian, Huawei and Lin, Weiguo},
 journal={Journal of Visual Communication and Image Representation},
 volume={98},
 pages={103981},
 year={2024},
 publisher={Elsevier}
}

If you have any questions, please contact me([email protected]).