[IJCAI 2025] Dual-Perspective United Transformer for Object Segmentation in Optical Remote Sensing Images

Yanguang Sun, Jiexi Yan, Jianjun Qian, Chunyan Xu, Jian Yang, Lei Luo

Our work has been accepted for IJCAI 2025. The code has already been open sourced.

If you are interested in our work, please do not hesitate to contact us at [email protected] via email.

Segmentation results

We provide the segmentation results of the proposed DPU-Former model under in Optical Remote Sensing Images.

DPU-Former_IJCAI25_ORSIs [(https://pan.baidu.com/s/1o0_eXVHwg2MNqv60LjXQZw), PIN:22mk]

Expanded Application

We also present the predicted results of our DPU-Former model in Camouflaged Object Detection (COD), Salient Object Detection (SOD), and Polyp Segmentation (PS).

DPU-Former_IJCAI25_COD [(https://pan.baidu.com/s/1EYqw-ydMt8C-u14MXmX20w), PIN:gje3]

DPU-Former_IJCAI25_SOD [(https://pan.baidu.com/s/1w34wWr2kBfke6KzLwGt--w), PIN:uncc]

DPU-Former_IJCAI25_PS [(https://pan.baidu.com/s/1OnNghgdlnAEgmIunRCUHFA), PIN:487z]

Weights

If you want to train or test our DPU-Former method, you can use the following weights.

DPU-Former_encoder [(https://pan.baidu.com/s/1LjGShJ9UNAKwaXXVlMmM7Q), PIN:kgma].

DPU-Former_ORSSD [(https://pan.baidu.com/s/1K6CWdBi4gwqZcnAQTxhMeg), PIN:p1nq].

DPU-Former_EORSSD [(https://pan.baidu.com/s/1cd_gwAJfjL2RqhSSjQoAtg), PIN:t2sy].

DPU-Former_ORS-4199 [(https://pan.baidu.com/s/1EMcshaJHHgcCl9ElqrjvOg), PIN:8hs5].

Citation

If you use DPU-Former in your research or wish to refer to the baseline results, please use the following BibTeX entry.

@article{DPU-Former,
  title={Dual-Perspective United Transformer for Object Segmentation in Optical Remote Sensing Images},
  author={Sun, Yanguang and Yan, Jiexi and Qian, Jianjun and Xu, Chunyan and Yang, Jian and Luo, Lei},
  journal={arXiv preprint arXiv:2506.21866},
  year={2025}
}

@article{DPU-Former,
  title={Dual-Perspective United Transformer for Object Segmentation in Optical Remote Sensing Images},
  author={Sun, Yanguang and Yan, Jiexi and Qian, Jianjun and Xu, Chunyan and Yang, Jian and Luo, Lei},
  journal={International Joint Conference on Artificial Intelligence (IJCAI)},
  year={2025}
}