This is an open-source implementation of paper: Map2DFusion: Real-time Incremental UAV Image Mosaicing based on Monocular SLAM.
Website : http://zhaoyong.adv-ci.com/map2dfusion/
Video : https://www.youtube.com/watch?v=-kSTDvGZ-YQ
PDF : http://zhaoyong.adv-ci.com/Data/map2dfusion/map2dfusion.pdf
If you use this project for research, please cite our paper:
@CONFERENCE{zhaoyong2016Map2DFusion,
author={S. {Bu} and Y. {Zhao} and G. {Wan} and Z. {Liu}},
booktitle={2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
title={Map2DFusion: Real-time incremental UAV image mosaicing based on monocular SLAM},
year={2016},
volume={},
number={},
pages={4564-4571},
doi={10.1109/IROS.2016.7759672},
ISSN={2153-0866},
month={Oct}
}
Download the latest code with:
Git:
git clone https://github.com/zdzhaoyong/Map2DFusion
Warnning: Compilation with CUDA can be enabled after CUDA_PATH defined.
If you are using linux systems, it can be compiled with one command (tested on ubuntu 14.04):
cd Map2DFusion;make
Obtain the sample sequence and launch:
git clone https://github.com/zdzhaoyong/phantom3-village-kfs
./Map2DFusion DataPath=phantom3-village-kfs
More sequences can be downloaded at the NPU DroneMap Dataset.
If you have any issue compiling/running Map2DFusion or you would like to know anything about the code, please contact the authors:
Yong Zhao -> zd5945@126.com