MaDiNet: Mamba Diffusion Network for SAR Target Detection (IEEE TCSVT 2025)

👑MaDiNet is the first work to integrate Gamma distribution into diffusion model with Mamba for SAR target detection.

✨We hope she can inspire more friends to delve into this fascinating field.

🛠️ Updates (YY/MM/DD)

(2024/10) Code is released.
(2025/05/28) MaDiNet has her unique identity card! (doi: 10.1109/TCSVT.2025.3574657).
(2025/11/02) MaDiNet has her publication date! (IEEE TCSVT (Volume: 35, Issue: 11, November 2025)).
(2026/03/15) MaDiNet has her life trophy of ESI 1% highly cited papers.

🕸️ Dataset

SSDD [link]
SAR-ShipDataset [link]
SAR-AIRcraft1.0 [link]
OGSOD-1.0 [link]

📽️ Getting Started

The installation instruction and usage are in Getting Started with DiffusionDet.
The MambaSAR is inspired by MambaVision of NVlabs and LeapLabTHU. Thanks to NVlabs and LeapLabTHU.

*The overall repository style is highly borrowed from DifffusionDet. Thanks to Shoufa Chen.

🍭 GOOD NEW!!!

MaDiNet has entered the top 1% of ESI highly cited papers and obtained her life trophy on March, 12, 2026 🎉🎉🎉 <img width="1857" height="537" alt="41ac602c90446b5b24b47d8fd55d41d9" src="https://github.com/user-attachments/assets/370052b1-1180-4b63-8812-76b4c65a9532" />

License

This project is under the CC-BY-NC 4.0 license. See LICENSE for details.

💡 Citing MaDiNet and DiffDet4SAR

If you find MaDiNet and DiffDet4SAR helpful to your research or wish to refer to the baseline results published here, please use the following BibTeX entry🥰.

@ARTICLE{zhou2025madinet,
  author={Zhou, Jie and Liu, Yongxiang and Peng, Bowen and Liu, Li and Li, Xiang},
  journal={IEEE Transactions on Circuits and Systems for Video Technology}, 
  title={MaDiNet: Mamba Diffusion Network for SAR Target Detection}, 
  year={2025},
  volume={35},
  number={11},
  pages={10787-10800},
  doi={10.1109/TCSVT.2025.3574657}}

@ARTICLE{10494361,
  author={Zhou, Jie and Xiao, Chao and Peng, Bo and Liu, Zhen and Liu, Li and Liu, Yongxiang and Li, Xiang},
  journal={IEEE Geoscience and Remote Sensing Letters}, 
  title={DiffDet4SAR: Diffusion-Based Aircraft Target Detection Network for SAR Images}, 
  year={2024},
  volume={21},
  number={},
  pages={1-5},
  keywords={Aircraft;Object detection;Radar polarimetry;Feature extraction;Scattering;Noise;Convolution;Aircraft target detection;diffusion model;synthetic aperture radar (SAR)},
  doi={10.1109/LGRS.2024.3386020}}

Please light up the STAR⭐⭐⭐⭐⭐ to encourage more opensource on SAR image interpretations!🥰🥳🥂

MaDiNet

Install / Use

README