zilunzhang / SODA-mmrotate

OpenMMLab Rotated Object Detection Toolbox and Benchmark

Introduction

This project is based on the open source object detection toolbox MMRotate, please refer to Installation for installation instructions first.

The benchmark experiments work with Python 3.8, PyTorch 1.10 and mmrotate 0.3.0, and corresponding configs can be found at sodaa-benchmarks.

Data preparation

Please refer to tools/data/sodaa for more details.

Citation

If you use our benchmark in your research, please cite this project.

@article{SODA,
  title={Towards large-scale small object detection: Survey and benchmarks},
  author={Cheng, Gong and Yuan, Xiang and Yao, Xiwen and Yan, Kebing and Zeng, Qinghua and Han, Junwei},
  journal={arXiv preprint arXiv:2207.14096},
  year={2022}
}

@inproceedings{zhou2022mmrotate,
  title   = {MMRotate: A Rotated Object Detection Benchmark using PyTorch},
  author  = {Zhou, Yue and Yang, Xue and Zhang, Gefan and Wang, Jiabao and Liu, Yanyi and
             Hou, Liping and Jiang, Xue and Liu, Xingzhao and Yan, Junchi and Lyu, Chengqi and
             Zhang, Wenwei and Chen, Kai},
  booktitle={Proceedings of the 30th ACM International Conference on Multimedia},
  year={2022}
}