FCOSR: A Simple Anchor-free Rotated Detector for Aerial Object Detection
arXiv preprint (arXiv). Article version RS

This implement is modified from mmdetection. We also refer to the codes of ReDet, PIoU, and ProbIoU.

In the process of implementation, we find that only Python code processing will produce huge memory overhead on Nvidia devices. Therefore, we directly write the label assignment module proposed in this paper in the form of CUDA extension of Pytorch. The program could not work effectively when we migrate it to cuda 11 (only support cuda10). By applying CUDA expansion, the memory utilization is improved and a lot of unnecessary calculations are reduced. We also try to train FCOSR-M on 2080ti (4 images per device), which can basically fill memory of graphics card.

FCOSR TensorRT inference code is available at: https://github.com/lzh420202/TensorRT_Inference

We add a multiprocess version DOTA2COCO into DOTA_devkit package, you could switch USE_MULTI_PROCESS to control the function in prepare_dota.py

Please refer to install.md for installation and dataset preparation.

Please see get_started.md for the basic usage.

Details (Test device: nvidia RTX 2080ti)

The password of baiduPan is ABCD

A part of Dota1.0 dataset (whole image mode) Code

@Article{rs15235499,
AUTHOR = {Li, Zhonghua and Hou, Biao and Wu, Zitong and Ren, Bo and Yang, Chen},
TITLE = {FCOSR: A Simple Anchor-Free Rotated Detector for Aerial Object Detection},
JOURNAL = {Remote Sensing},
VOLUME = {15},
YEAR = {2023},
NUMBER = {23},
ARTICLE-NUMBER = {5499},
URL = {https://www.mdpi.com/2072-4292/15/23/5499},
ISSN = {2072-4292},
DOI = {10.3390/rs15235499}
}