r3det-pytorch

KO | EN

gitlite — search

#typescript #ai-agents #ai #dsh-plugin #deepseek-harness #open-source #cli #claude-code #codex #developer-tools #react #windows

★ 44 Open GitHub ↗

R3Det based on mmdet 2.19.0

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object

[![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)
[![arXiv](http://img.shields.io/badge/cs.CV-arXiv%3A1908.05612-B31B1B.svg)](https://arxiv.org/abs/1908.05612)

## News
2022.02.18 MMRotate: OpenMMLab Rotated Object Detection Benchmark was released!

## Installation

```shell
# install mmdetection first if you haven't installed it yet. (Refer to mmdetection for details.)
pip install mmdet==2.19.0

# install r3det (Compiling rotated ops is a little time-consuming.)
pip install -r requirements.txt
pip install -v -e .

```

- It is best to use opencv-python greater than 4.5.1 because its angle representation has been changed in 4.5.1. The following experiments are all run with 4.5.3.

## Quick Start

Please change [path](configs/_base_/datasets/dota1_0.py#L3) in configs to your data path.
```shell
# train
CUDA_VISIBLE_DEVICES=0 PORT=29500 \
./tools/dist_train.sh configs/rretinanet/rretinanet_obb_r50_fpn_1x_dota_v3.py 1

# submission
CUDA_VISIBLE_DEVICES=0 PORT=29500 \
./tools/dist_test.sh configs/rretinanet/rretinanet_obb_r50_fpn_1x_dota_v3.py \
        work_dirs/rretinanet_obb_r50_fpn_1x_dota_v3/epoch_12.pth 1 --format-only\
        --eval-options submission_dir=work_dirs/rretinanet_obb_r50_fpn_1x_dota_v3/Task1_results
```

For DOTA dataset, please crop the original images into 1024×1024 patches with an overlap of 200 by run
```shell
python tools/split/img_split.py --base_json \
       tools/split/split_configs/dota1_0/ss_trainval.json

python tools/split/img_split.py --base_json \
       tools/split/split_configs/dota1_0/ss_test.json

```
Please change path in [ss_trainval.json](./tools/split/split_configs/dota1_0/ss_trainval.json#L4-L11), [ss_test.json](./tools/split/split_configs/dota1_0/ss_test.json#L5)  to your path. (Forked from [BboxToolkit](https://github.com/jbwang1997/BboxToolkit), which is faster then DOTA_Devkit.)

### Angle Representations
Three angle representations are built-in, which can freely switch in the config.

- `v1` (from R3Det): [-PI/2, 0)
- `v2` (from S2ANet): [-Pi/4, 3PI/4)
- `v3` (from OBBDetection): [-PI/2, PI/2)

The differences of the three angle representations are reflected in poly2obb, obb2poly, obb2xyxy, obb2hbb, hbb2obb, etc. [[More](./r3det/core/bbox/rtransforms.py)], And according to the above three papers, the coders of them are different.

- DeltaXYWHAOBBoxCoder
  - `v1`：None
  - `v2`：Constrained angle + Projection of dx and dy + Normalized with PI
  - `v3`：Constrained angle and length&width + Projection of dx and dy
- DeltaXYWHAHBBoxCoder
  - `v1`：None
  - `v2`：Constrained angle  + Normalized with PI
  - `v3`：Constrained angle and length&width  + Normalized with 2PI

**We believe that different coders are the key reason for the different baselines in different papers.** The good news is that all the above coders can be freely switched in R3Det. In addition, R3Det also provide 4 NMS ops and 3 IoU_Calculators for rotation detection as follows:

- `nms.type`
  - v1：`v1`
  - v2：`v2`
  - v3：`v3`
  - mmcv: `mmcv`
- `iou_calculator`
  - v1：`RBboxOverlaps2D_v1`
  - v2：`RBboxOverlaps2D_v2`
  - v3：`RBboxOverlaps2D_v3`

### Performance

| Model | Backbone | Lr schd | MS | RR | Angle | box AP | Official | Download |
|:--------:|:--------:|:-------:|:--:|:------:|:--------:|:------:|:------:|:------:|
|RRetinaNet HBB | R50-FPN | 1x | - | - | v1 | 65.19 | [65.73](https://github.com/yangxue0827/RotationDetection) | [Baidu:0518](https://pan.baidu.com/s/1ijkb0y_yAaicT-Z9_ljKeA)/[Google](https://drive.google.com/drive/folders/1CeD3QPTQRRSI7WKMwWE3EUWhzD2qN4e4?usp=sharing)
|RRetinaNet OBB| R50-FPN | 1x | - | - | v3 | 68.20 | [69.40](https://github.com/jbwang1997/OBBDetection/tree/master/configs/obb/retinanet_obb) | [Baidu:0518](https://pan.baidu.com/s/1ijkb0y_yAaicT-Z9_ljKeA)/[Google](https://drive.google.com/drive/folders/1CeD3QPTQRRSI7WKMwWE3EUWhzD2qN4e4?usp=sharing) |
|RRetinaNet OBB | R50-FPN | 1x | - | - | v2 | 68.64 | [68.40](https://github.com/csuhan/s2anet) | [Baidu:0518](https://pan.baidu.com/s/14o4sNxzfWQj1oGFjBzX8Kg)/[Google]()|
|R3Det| R50-FPN | 1x | - | - | v1 | 70.41 | [70.66](https://github.com/yangxue0827/RotationDetection) | [Baidu:0518](https://pan.baidu.com/s/1ECNAzE3xaXXO7Pj2p_bLDw)/[Google]() |
|R3Det*| R50-FPN | 1x | - | - | v1 | 70.86 | - | [Baidu:0518](https://pan.baidu.com/s/1kWg-bz2KjDcI-s_IWvUE6A)/[Google]() |

- `MS` means multiple scale image split.
- `RR` means random rotation.

## Citation

```
@inproceedings{yang2021r3det,
    title={R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object},
    author={Yang, Xue and Yan, Junchi and Feng, Ziming and He, Tao},
    booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
    volume={35},
    number={4},
    pages={3163--3171},
    year={2021}
}

```

r3det-pytorch

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README