https://github.com/xiaoyan07/SAM_MLoRA

https://github.com/xiaoyan07/SAM_MLoRA

Last synced: about 1 month ago
JSON representation

• Host: GitHub
• URL: https://github.com/xiaoyan07/SAM_MLoRA
• Owner: xiaoyan07
• License: mit
• Created: 2024-04-23T08:21:50.000Z (9 months ago)
• Default Branch: main
• Last Pushed: 2024-09-12T11:42:13.000Z (4 months ago)
• Last Synced: 2024-09-12T22:58:29.101Z (4 months ago)
• Language: Python
• Size: 2.25 MB
• Stars: 2
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• Awesome-Segment-Anything - [code

README

        
Multi-LoRA Fine-Tuned Segment Anything Model for Extraction of Urban Man-Made Objects 


 Xiaoyan LU and

Qihao WENG



[[`Paper`](https://ieeexplore.ieee.org/abstract/document/10637992)] 

## Multi-LoRA Fine-Tuned SAM Framework



  



## The training dataset

1. [DeepGlobe Road Training Dataset ](https://competitions.codalab.org/competitions/18467#participate-get_data): 4696 samples

2. [SpaceNet Building AOI2 and AOI4 Dataset ](https://spacenet.ai/spacenet-buildings-dataset-v2/): 8429 samples

## The validation set

1. [DeepGlobe Road Test Dataset ](https://competitions.codalab.org/competitions/18467#participate-get_data): 1530 samples

2. [SpaceNet Building AOI3 and AOI5 Dataset ](https://spacenet.ai/spacenet-buildings-dataset-v2/): 1148 (Paris) and 1101 (Khartoum) samples

3. [The WHU building (Christchurch) dataset](http://gpcv.whu.edu.cn/data/building_dataset.html): 2416 samples

##

2. The trained weights of SAM_Adapter, SAM_LoRA (r=96), and SAM_MLoRA (r=32,n=3) are released at [Baidu Drive](https://pan.baidu.com/s/1_uscV5L8mDt4kihwt1iSqg), Code: MODE

## Road Extraction

SAM_Adapter

```

python train_sam_adapter.py --name='b_adapter_sam'

```

SAM_LoRA (r=96) 

```

python train_sam_adapter.py --name='b_adapter_sam_lora96_96'

```

SAM_MLoRA (r=32,n=3)

```

python train_sam_adapter.py --name='b_adapter_sam_multi_lora'

```

## Building Extraction

SAM_Adapter

```

python train_sam_adapter_build.py --name='b_adapter_sam_sp24'

```

SAM_LoRA (r=96) 

```

python train_sam_adapter_build.py --name='b_adapter_sam_lora96_96_sp24'

```

SAM_MLoRA (r=32,n=3)

```

python train_sam_adapter_build.py --name='b_adapter_sam_multi_lora32_sp24'

```

## Citation

If this code or dataset contributes to your research, please kindly consider citing our paper :)

```

@article{Lu2024MLoRA,

    title = {Multi-LoRA Fine-Tuned Segment Anything Model for Urban Man-Made Object Extraction},

    author = {Xiaoyan LU and Qihao Weng},

    journal = {IEEE Transactions on Geoscience and Remote Sensing},

    volume = {62},

    pages = {1-19},

    year = {2024},

    doi = {https://doi.org/10.1109/TGRS.2024.3435745}

}

```