BDCL-Net Prediction Maps

This repository provides the prediction maps of our paper:

Bilateral Decoupling Complementarity Learning Network for Camouflaged Object Detection

Rui Zhao, Yuetong Li, Qing Zhang, and Xinyi Zhao

Knowledge-Based Systems, Vol. 314, Article 113158, 2025.

DOI: 10.1016/j.knosys.2025.113158

## Quantitative Results

The quantitative performance of BDCL-Net on four benchmark datasets is summarized below.

## Prediction Maps

We provide the prediction maps of BDCL-Net on four commonly used camouflaged object detection benchmarks:

• CAMO
• COD10K
• NC4K
• CHAMELEON

The files are organized as follows:

Prediction_Maps/
├── CAMO/
├── COD10K/
├── NC4K/
└── CHAMELEON/

Each folder contains the predicted camouflage maps corresponding to the test images of the respective dataset.

These prediction maps can be directly used for quantitative evaluation and qualitative comparison with other camouflaged object detection methods.

## Citation

If you find these prediction maps useful for your research, please cite our paper:

@article{zhao2025bilateral,
  title={Bilateral decoupling complementarity learning network for camouflaged object detection},
  author={Zhao, Rui and Li, Yuetong and Zhang, Qing and Zhao, Xinyi},
  journal={Knowledge-Based Systems},
  volume={314},
  pages={113158},
  year={2025},
  publisher={Elsevier},
  doi={10.1016/j.knosys.2025.113158}
}