Partial Weakly-Supervised Oriented Object Detection

Mingxin Liu  Peiyuan Zhang  Yuan Liu  Wei Zhang  Yue Zhou  Ning Liao  Ziyang Gong  Junwei Luo  Zhirui Wang  Yi Yu  Xue Yang 

Introduction

We propose the first Partial Weakly-Supervised Oriented Object Detection (PWOOD) framework based on partially weak annotations (horizontal boxes or single points), which can efficiently leverage large amounts of unlabeled data, significantly outperforming weakly supervised algorithms trained with partially weak annotations, and also offers a lower cost solution.

Installation

conda create -n mm python==3.8 -y
conda activate mm

pip install torch==1.13.1+cu116 torchvision==0.14.1+cu116 torchaudio==0.13.1 --extra-index-url https://download.pytorch.org/whl/cu116
pip install -U openmim
mim install mmcv-full
mim install mmdet\<3.0.0

pip install scikit-learn
pip install prettytable

# For Point branch
cd mmrotate
pip install -v -e .

Data Preparation

DOTA

1. Labeled/Unlabeled Data Division

To divide the DOTA- v1.0/v1.5 dataset into labeled and unlabeled data, please refer to Data preparation of SOOD .

To divide the DOTA- v2.0 into labeled and unlabeled data, please refer to data_list/dotav2.

2. Data Split

For details on how to split the DOTA dataset into patches, please refer to the official implementation .

After split, the data folder should be organized as follows:

split_ss_dota_vxx
├── train
│   ├── images
│   └── annfiles
├── val
│   ├── images
│   └── annfiles
├── train_xx_labeled
│   ├── images
│   └── annfiles
└──train_xx_unlabeled
    ├── images
    └── annfiles

DIOR

To divide the DIOR into labeled and unlabeled data, please refer to data_list/dior.

Train

#2 GPU
CUDA_VISIBLE_DEVICES=0,1 python -m torch.distributed.launch --nnodes=1 \
--node_rank=0 --master_addr="127.0.0.1" --nproc_per_node=2 --master_port=25510 \
train.py configs_dota15/xxx/xxx.py \
--launcher pytorch \
--work-dir work_dir/xxx/

Test

python test.py configs_dota15/xxx/xxx.py work_dir/xxx/xxx.pth 

Weight

DOTA- v1.0

Labeled Data	mAP	Config	Model	Log
20%	62.93	semi_h2rv2_adamw_dotav1_20p.py	best_0.629314_mAP.pth	dotav1_20p_log
30%	65.42	semi_h2rv2_adamw_dotav1_30p.py	best_0.654153_mAP.pth	dotav1_30p_log

DOTA- v1.5

Labeled Data	mAP	Config	Model	Log
10%	52.87	semi_h2rv2_adamw_dota15_10p.py	best_0.528748_mAP.pth	dotav15_10p_log
20%	59.36	semi_h2rv2_adamw_dota15_20p.py	best_0.593614_mAP.pth	dotav15_20p_log
30%	61.58	semi_h2rv2_adamw_dota15_30p.py	best_0.615836_mAP.pth	dotav15_30p_log

DOTA- v2.0

Labeled Data	mAP	Config	Model	Log
10%	31.30	semi_h2rv2_adamw_dota2_10p.py	best_0.310266_mAP.pth	dotav2_10p_log
20%	36.39	semi_h2rv2_adamw_dota2_20p.py	best_0.363926_mAP.pth	dotav2_20p_log
30%	40.27	semi_h2rv2_adamw_dota2_30p.py	best_0.402659_mAP.pth	dotav2_30p_log

DIOR

Labeled Data	mAP	Config	Model	Log
10%	54.33	semi_h2rv2_adamw_dior_10p.py	best_0.543296_mAP.pth	doir_10p_log
20%	57.89	semi_h2rv2_adamw_dior_20p.py	best_0.578923_mAP.pth	dior_20p_log
30%	60.42	semi_h2rv2_adamw_dior_30p.py	best_0.604248_mAP.pth	dior_30p_log

Guide

If you need Point version, please switch to the Point branch.

Citation

@misc{liu2025partialweaklysupervisedorientedobject,
      title={Partial Weakly-Supervised Oriented Object Detection}, 
      author={Mingxin Liu and Peiyuan Zhang and Yuan Liu and Wei Zhang and Yue Zhou and Ning Liao and Ziyang Gong and Junwei Luo and Zhirui Wang and Yi Yu and Xue Yang},
      year={2025},
      eprint={2507.02751},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2507.02751}, 
}