yxhnjust / F3Net

The code of the paper F3-Net: Feature Fusion and Filtration Network for Object Detection in Optical Remote Sensing Images.

F3-Net

Abstract

The code of the paper F3-Net: Feature Fusion and Filtration Network for Object Detection in Optical Remote Sensing Images. The method is based on Faster R-CNN and is completed by Xinhai Ye.

Environment

1. python 3.5
2. cuda 10.0
3. opencv
4. tensorflow-gpu 1.13
5. tfplot 0.2.0

Hardware

At least two NVIDIA GPUs with more than 10GB memory.

Installation

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

Citation

If you find our work or code useful in your research, please consider citing:

@Article{rs12244027,
AUTHOR = {Ye, Xinhai and Xiong, Fengchao and Lu, Jianfeng and Zhou, Jun and Qian, Yuntao},
TITLE = {ℱ3-Net: Feature Fusion and Filtration Network for Object Detection in Optical Remote Sensing Images},
JOURNAL = {Remote Sensing},
VOLUME = {12},
YEAR = {2020},
NUMBER = {24},
ARTICLE-NUMBER = {4027},
ISSN = {2072-4292},
}