Awesome

SARNet_RS21

Code and result repository for our paper "Semantic-Guided Attention Refinement Network for Salient Object Detection in Optical Remote Sensing Images."

0. Prerequisites

Note that SARNet is only tested on Win_OS with the following environments. It may work on other operating systems as well, but we do not guarantee that it will.

○ Creating a virtual environment in terminal: conda create -n SARNet python=3.6.

○ Installing necessary packages: pip install -r requirements.txt.

1. Download training and test sets

Download the ORSSD and EORSSD datasets at the following link address:

○ EORSSD dataset

○ ORSSD dataset

2. Results Download

The training and testing experiments are conducted using PyTorch with a single GeForce RTX 3090 GPU of 24 GB Memory.

Our SARNet test results on ORSSD and EORSSD datasets:

○ BaiduYun (code: akyu)

○ Google Drive

3. Evaluation

You can evaluate the result maps using the tool in Matlab Version or Python_GPU Version.

4. Citation

Please cite our paper if you find the work useful:

@article{huang2021semantic,
  title={Semantic-Guided Attention Refinement Network for Salient Object Detection in Optical Remote Sensing Images},
  author={Huang, Zhou and Chen, Huaixin and Liu, Biyuan and Wang, Zhixi},
  journal={Remote Sensing},
  volume={13},
  number={11},
  pages={2163},
  year={2021},
  publisher={Multidisciplinary Digital Publishing Institute}
}

5. Acknowledgement

Thanks to Deng-Ping Fan, Yu-Wei Jin, and Run-Min Cong for their help in our work.