---
license: cc-by-nc-sa-4.0
---
# Model Card for Oriented R-CNN pretrained on DOTA 1.0

<!-- Provide a quick summary of what the model is/does. [Optional] -->
The original paper is [Oriented R-CNN for Object Detection](https://openaccess.thecvf.com/content/ICCV2021/papers/Xie_Oriented_R-CNN_for_Object_Detection_ICCV_2021_paper.pdf). 

This implementation of this model has been developed by [OpenMMLab](https://openmmlab.com/) in the [MMRotate](https://github.com/open-mmlab/mmrotate) framework.

The model has been trained on [DOTA 1.0](https://captain-whu.github.io/DOTA/)

The performance measured as mAP is 75.69.

- **Developed by:** OpenMMLab
- **Model type:** Object Detection model
- **License:** cc-by-nc-sa-4.0
- **Resources for more information:** More information needed
    - [GitHub Repo](https://github.com/open-mmlab/mmrotate/)
    - [Associated Paper](https://openaccess.thecvf.com/content/ICCV2021/papers/Xie_Oriented_R-CNN_for_Object_Detection_ICCV_2021_paper.pdf)


# How to Get Started with the Model

Use the code below to get started with the model.

```
from mmdet.apis import init_detector, inference_detector
import mmrotate

config_file = 'oriented_rcnn_r50_fpn_1x_dota_le90.py'
checkpoint_file = 'oriented_rcnn_r50_fpn_1x_dota_le90-6d2b2ce0.pth'
model = init_detector(config_file, checkpoint_file, device='cuda:0')
inference_detector(model, 'demo/demo.jpg')
```