Arbitrary-Oriented Object Detection with Circular Smooth Label

Abstract

This repo is based on Focal Loss for Dense Object Detection, and it is completed by YangXue.

Other CSL-based code: R³Det-CSL, OHDet

Pipeline

Circular Smooth Label

Latest Performance

DOTA1.0 (Task1)

Model	Backbone	Training data	Val data	mAP	Model Link	Anchor	Label Mode	Reg. Loss	Angle Range	lr schd	Data Augmentation	GPU	Image/GPU	Configs
CSL	ResNet50_v1 600->800	DOTA1.0 trainval	DOTA1.0 test	67.38	-	H	Gaussian (r=1, w=10)	smooth L1	180	2x	×	3X GeForce RTX 2080 Ti	1	cfgs_res50_dota_v45.py
CSL	ResNet50_v1 600->800	DOTA1.0 trainval	DOTA1.0 test	68.73	-	H	Pulse (w=1)	smooth L1	180	2x	×	2X GeForce RTX 2080 Ti	1	cfgs_res50_dota_v41.py

Notice:
Due to the improvement of the code, the performance of this repo is gradually improving, so the experimental results in other configuration files are for reference only.
Please refer to new repo for the latest progress.

Visualization

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 1.13

Download Model

Pretrain weights

1、Please download resnet50_v1, resnet101_v1 pre-trained models on Imagenet, put it to data/pretrained_weights.
2、(Recommend) Or you can choose to use a better backbone, refer to gluon2TF.

• Baidu Drive, password: 5ht9.
• Google Drive

Compile

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

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

Train

1、If you want to train your own data, please note:

(1) Modify parameters (such as CLASS_NUM, DATASET_NAME, VERSION, etc.) in $PATH_ROOT/libs/configs/cfgs.py
(2) Add category information in $PATH_ROOT/libs/label_name_dict/label_dict.py     
(3) Add data_name to $PATH_ROOT/data/io/read_tfrecord.py 

2、Make tfrecord
For DOTA dataset:

cd $PATH_ROOT\data\io\DOTA
python data_crop.py

cd $PATH_ROOT/data/io/  
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、Multi-gpu train

cd $PATH_ROOT/tools
python multi_gpu_train.py

Eval

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

Tensorboard

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

Citation

If this is useful for your research, please consider cite.

@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}
}

@inproceedings{xia2018dota,
    title={DOTA: A large-scale dataset for object detection in aerial images},
    author={Xia, Gui-Song and Bai, Xiang and Ding, Jian and Zhu, Zhen and Belongie, Serge and Luo, Jiebo and Datcu, Mihai and Pelillo, Marcello and Zhang, Liangpei},
    booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
    pages={3974--3983},
    year={2018}
}

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