YichengQiao / SeMask-Mask2Former

SeMask-Mask2Former: A Semantic Segmentation Model for High Resolution Remote Sensing Images

English Version:

SeMask-Mask2Former：A Semantic Segmentation Model for High Resolution Remote Sensing Images

This project presents a semantic segmentation model for high-resolution remote sensing images called SeMask-Mask2Former, based on the potsdam dataset. Below is the directory structure and a brief description of the project.

File and Folder Descriptions

• .ipynb_checkpoints: Jupyter Notebook checkpoint folder.
• configs: Contains model configuration files.
• data: Contains dataset and preprocessing-related files.
• datasets: Contains code related to dataset processing.
• demo: Contains demo files.
• output: Contains output results from training and testing.
• tools: Contains auxiliary tools.
• eval.sh: Evaluation script.
• GETTING_STARTED.md: Getting started guide.
• INSTALL.md: Installation guide.
• kill_inpynb.checkpoint.py: Jupyter Notebook processing file.
• mmcv.ipynb: MMCV library usage example.
• MODEL_ZOO.md: Model zoo description.
• README.md: Project readme file.
• register_data.ipynb: Dataset registration file.
• requirements.txt: List of Python packages required for the project.
• test_vaihingen.ipynb: Test file for the Vaihingen dataset.
• train.ipynb: Jupyter Notebook file for training.
• train.sh: Training script.
• train_ipynb.txt: Training text file.
• train_net.py: Main code for network training.
• train_potsdam-old.ipynb: Training file for the old version of the Potsdam dataset.
• train_vaihingen.ipynb: Training file for the Vaihingen dataset.

Quick Start

Please refer to INSTALL.md and GETTING_STARTED.md for project installation and configuration.

Model Training and Testing

Use train.ipynb, train_vaihingen.ipynb, and train_potsdam-old.ipynb for model training. For testing, refer to test_vaihingen.ipynb.

Enjoy using the project!

citation

If you find this project helpful for your research, please consider citing the report and giving a ⭐.

@INPROCEEDINGS{10115761,
  author={Qiao, Yicheng and Liu, Wei and Liang, Bin and Wang, Pengyun and Zhang, Haopeng and Yang, Junli},
  booktitle={2023 IEEE Aerospace Conference}, 
  title={SeMask-Mask2Former: A Semantic Segmentation Model for High Resolution Remote Sensing Images}, 
  year={2023},
  volume={},
  number={},
  pages={1-6},
  doi={10.1109/AERO55745.2023.10115761}
}

Contribute

I am grateful for the contributions made by Wei Liu, Bin Liang, and Pengyun Wang towards this project. This is our joint work.

中文版本：

SeMask-Mask2Former：A Semantic Segmentation Model for High Resolution Remote Sensing Images

这个项目是一个基于potsdam数据集的遥感图像语义分割模型，名为SeMask-Mask2Former。以下是项目的目录结构和简要说明。

*《SeMask-Mask2Former：A Semantic Segmentation Model for High Resolution Remote Sensing Images》发表在2023 IEEE Aerospace Conference At the Yellowstone Conference Center in Big Sky, Montana,March 04 - 11 .

├─ .ipynb_checkpoints

├─ configs

├─ data

├─ datasets

├─ demo

├─ output

├─ tools

├─ eval.sh

├─ GETTING_STARTED.md

├─ INSTALL.md

├─ kill_inpynb.checkpoint.py

├─ mmcv.ipynb

├─ MODEL_ZOO.md

├─ README.md

├─ register_data.ipynb

├─ requirements.txt

├─ test_vaihingen.ipynb

├─ train.ipynb

├─ train.sh

├─ train_ipynb.txt

├─ train_net.py

├─ train_potsdam-old.ipynb

└─ train_vaihingen.ipynb

文件及文件夹说明

• .ipynb_checkpoints: Jupyter Notebook的检查点文件夹。
• configs: 存放模型配置文件。
• data: 存放数据集和预处理过程的相关文件。
• datasets: 包含数据集处理的相关代码。
• demo: 存放演示文件。
• output: 存放训练和测试的输出结果。
• tools: 一些辅助工具。
• eval.sh: 评估脚本。
• GETTING_STARTED.md: 入门指南。
• INSTALL.md: 安装指南。
• kill_inpynb.checkpoint.py: Jupyter Notebook处理文件。
• mmcv.ipynb: MMCV库的使用示例。
• MODEL_ZOO.md: 模型库说明。
• README.md: 项目的自述文件。
• register_data.ipynb: 数据集注册文件。
• requirements.txt: 项目依赖的Python包列表。
• test_vaihingen.ipynb: Vaihingen数据集的测试文件。
• train.ipynb: 训练的Jupyter Notebook文件。
• train.sh: 训练脚本。
• train_ipynb.txt: 训练文本文件。
• train_net.py: 训练网络的主要代码。
• train_potsdam-old.ipynb: 旧版Potsdam数据集的训练文件。
• train_vaihingen.ipynb: Vaihingen数据集的训练文件。

快速开始

请参照INSTALL.md和GETTING_STARTED.md进行项目的安装和配置。

安装

git clone https://github.com/Joeyicheng/SeMask-Mask2Former.git

pip install -r requirements.txt

模型训练与测试

使用train.ipynb、train_vaihingen.ipynb和train_potsdam-old.ipynb进行模型的训练。测试过程可参考test_vaihingen.ipynb。

祝您使用愉快！

引用

如果你觉得这个项目对你的研究有帮助，请考虑引用报告，并给出一个 ⭐.

@INPROCEEDINGS{10115761,
  author={Qiao, Yicheng and Liu, Wei and Liang, Bin and Wang, Pengyun and Zhang, Haopeng and Yang, Junli},
  booktitle={2023 IEEE Aerospace Conference}, 
  title={SeMask-Mask2Former: A Semantic Segmentation Model for High Resolution Remote Sensing Images}, 
  year={2023},
  volume={},
  number={},
  pages={1-6},
  doi={10.1109/AERO55745.2023.10115761}
}