UranusDet

Abstract

This is a tensorflow-based rotation detection benchmark, also called UranusDet. UranusDet is completed by YangXue.

Papers and codes related to remote sensing/aerial image detection: DOTA-DOAI.

Techniques:

• Dataset: DOTA, HRSC2016, ICDAR2015, ICDAR2017 MLT, UCAS-AOD, FDDB, OHD-SJTU, SSDD++
• Baackbone: ResNet, MobileNetV2, EfficientNet, DarkNet53
• Neck: FPN, BiFPN
• Detectors:
  • R²CNN (Faster-RCNN-H): TF code
  • RRPN (Faster-RCNN-R): TF code
  • SCRDet: R²CNN++, IoU-Smooth L1 Loss
  • RetinaNet-H, RetinaNet-R: TF code
  • RefineRetinaNet (CascadeRetinaNet): TF code
  • FCOS
  • RSDet
  • R³Det: TF code, Pytorch code
  • Circular Smooth Label (CSL): TF code
  • Densely Coded Label (DCL): TF code
  • Mixed method: R³Det-DCL
• Loss: CE, Focal Loss, Smooth L1 Loss, IoU-Smooth L1 Loss, Modulated Loss

Projects

Latest Performance

More results and trained models are available in the MODEL_ZOO.md.

DOTA1.0 (Task1)

Model	Neck	Backbone	Training/test dataset	mAP	Model Link	Anchor	Angle Pred.	Reg. Loss	Angle Range	Data Augmentation	Configs
RetinaNet-H	FPN	ResNet50_v1d 600->800	DOTA1.0 trainval/test	64.17	Baidu Drive (j5l0)	H	Reg.	smooth L1	180	×	cfgs_res50_dota_v15.py
RetinaNet-H	FPN	ResNet50_v1d 600->800	DOTA1.0 trainval/test	65.73	Baidu Drive (jum2)	H	Reg.	smooth L1	90	×	cfgs_res50_dota_v4.py
RSDet	FPN	ResNet50_v1d 600->800	DOTA1.0 trainval/test	67.27	Baidu Drive (6nt5)	H	Reg.	modulated loss	-	×	cfgs_res50_dota_rsdet_v2.py
CSL	FPN	ResNet50_v1d 600->800	DOTA1.0 trainval/test	67.38	Baidu Drive (g3wt)	H	Cls.: Gaussian (r=1, w=10)	smooth L1	180	x	cfgs_res50_dota_v45.py
DCL	FPN	ResNet50_v1d 600->800	DOTA1.0 trainval/test	67.39	Baidu Drive (p9tu)	H	Cls.: BCL (w=180/256)	smooth L1	180	×	cfgs_res50_dota_dcl_v5.py
R3Det	FPN	ResNet50_v1d 600->800	DOTA1.0 trainval/test	70.66	Baidu Drive (30lt)	H->R	Reg.	smooth L1	90	×	cfgs_res50_dota_r3det_v1.py
R3Det-DCL	FPN	ResNet50_v1d 600->800	DOTA1.0 trainval/test	71.21	Baidu Drive (jueq)	H->R	Cls.: BCL (w=180/256)	iou-smooth L1	90->180	×	cfgs_res50_dota_r3det_dcl_v1.py
R2CNN (Faster-RCNN)	FPN	ResNet50_v1d 600->800	DOTA1.0 trainval/test	72.27	Baidu Drive (wt2b)	H->R	Reg.	smooth L1	90	×	cfgs_res50_dota_v1.py

My Development Environment

docker images: docker pull yangxue2docker/yx-tf-det:tensorflow1.13.1-cuda10-gpu-py3

1. python3.5 (anaconda recommend)
2. cuda 10.0
3. opencv(cv2)
4. tfplot 0.2.0 (optional)
5. tensorflow-gpu 1.13

Download Model

Pretrain weights

1. Please download resnet50_v1, resnet101_v1, resnet152_v1, efficientnet, mobilenet_v2, darknet53 (Baidu Drive (1jg2), Google Drive) pre-trained models on Imagenet, put them to $PATH_ROOT/dataloader/pretrained_weights.
2. (Recommend in this repo) Or you can choose to use better backbones (resnet_v1d), refer to gluon2TF.

• Baidu Drive (5ht9)
• Google Drive

Trained weights

1. Please download trained models by this project, then put them to $PATH_ROOT/output/pretained_weights.

Compile

```  
cd $PATH_ROOT/libs/utils/cython_utils
python setup.py build_ext --inplace (or make)

cd $PATH_ROOT/libs/utils/
python setup.py build_ext --inplace
```

Train

1. If you want to train your own dataset, please note:

   (1) Select the detector and dataset you want to use, and mark them as #DETECTOR and #DATASET (such as #DETECTOR=retinanet and #DATASET=DOTA)
   (2) Modify parameters (such as CLASS_NUM, DATASET_NAME, VERSION, etc.) in $PATH_ROOT/libs/configs/#DATASET/#DETECTOR/cfgs_xxx.py
   (3) Copy $PATH_ROOT/libs/configs/#DATASET/#DETECTOR/cfgs_xxx.py to $PATH_ROOT/libs/configs/cfgs.py
   (4) Add category information in $PATH_ROOT/libs/label_name_dict/label_dict.py     
   (5) Add data_name to $PATH_ROOT/data/io/read_tfrecord.py  
   

2. Make tfrecord
   If image is very large (such as DOTA dataset), the image needs to be cropped. Take DOTA dataset as a example:

   cd $PATH_ROOT/dataloader/dataset/DOTA
   python data_crop.py
   

   If image does not need to be cropped, just convert the annotation file into xml format, refer to example.xml.

   cd $PATH_ROOT/dataloader/dataset/  
   python convert_data_to_tfrecord.py --VOC_dir='/PATH/TO/DOTA/' 
                                      --xml_dir='labeltxt'
                                      --image_dir='images'
                                      --save_name='train' 
                                      --img_format='.png' 
                                      --dataset='DOTA'
   

3. Start training

   cd $PATH_ROOT/tools/#DETECTOR
   python train.py
   

Test

1. For large-scale image, take DOTA dataset as a example (the output file or visualization is in $PATH_ROOT/tools/#DETECTOR/test_dota/VERSION):

   cd $PATH_ROOT/tools/#DETECTOR
   python test_dota_ms.py --test_dir='/PATH/TO/IMAGES/'  
                          --gpus=0,1,2,3,4,5,6,7  
                          -ms (multi-scale testing, optional)
                          -s (visualization, optional)
   

   Notice: In order to set the breakpoint conveniently, the read and write mode of the file is' a+'. If the model of the same #VERSION needs to be tested again, the original test results need to be deleted.

2. For small-scale image, take HRSC2016 dataset as a example:

   cd $PATH_ROOT/tools/#DETECTOR
   python test_hrsc2016_ms.py --test_dir='/PATH/TO/IMAGES/'  
                              --gpu=0
                              --image_ext='bmp'
                              --test_annotation_path='/PATH/TO/ANNOTATIONS'
                              -s (visualization, optional)
   

Tensorboard

cd $PATH_ROOT/output/summary
tensorboard --logdir=.

Citation

If you find our code useful for your research, please consider cite.

@article{yang2020dense,
    title={Dense Label Encoding for Boundary Discontinuity Free Rotation Detection},
    author={Yang, Xue and Hou, Liping and Zhou, Yue and Wang, Wentao and Yan, Junchi},
    journal={arXiv preprint arXiv:2011.09670},
    year={2020}
}

@article{yang2020arbitrary,
    title={Arbitrary-Oriented Object Detection with Circular Smooth Label},
    author={Yang, Xue and Yan, Junchi},
    journal={European Conference on Computer Vision (ECCV)},
    year={2020}
    organization={Springer}
}

@article{yang2019r3det,
    title={R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object},
    author={Yang, Xue and Liu, Qingqing and Yan, Junchi and Li, Ang and Zhang, Zhiqiang and Yu, Gang},
    journal={arXiv preprint arXiv:1908.05612},
    year={2019}
}

@article{qian2019learning,
    title={Learning modulated loss for rotated object detection},
    author={Qian, Wen and Yang, Xue and Peng, Silong and Guo, Yue and Yan, Chijun},
    journal={arXiv preprint arXiv:1911.08299},
    year={2019}
}

@article{yang2020scrdet++,
    title={SCRDet++: Detecting Small, Cluttered and Rotated Objects via Instance-Level Feature Denoising and Rotation Loss Smoothing},
    author={Yang, Xue and Yan, Junchi and Yang, Xiaokang and Tang, Jin and Liao, Wenglong and He, Tao},
    journal={arXiv preprint arXiv:2004.13316},
    year={2020}
}

@inproceedings{yang2019scrdet,
    title={SCRDet: Towards more robust detection for small, cluttered and rotated objects},
    author={Yang, Xue and Yang, Jirui and Yan, Junchi and Zhang, Yue and Zhang, Tengfei and Guo, Zhi and Sun, Xian and Fu, Kun},
    booktitle={Proceedings of the IEEE International Conference on Computer Vision (ICCV)},
    pages={8232--8241},
    year={2019}
}

Reference

1、https://github.com/endernewton/tf-faster-rcnn
2、https://github.com/zengarden/light_head_rcnn
3、https://github.com/tensorflow/models/tree/master/research/object_detection
4、https://github.com/fizyr/keras-retinanet