This released code is the VIO part of VI-HSO based on HSO.

Author: Wenzhe Yang, Yan Zhuang, Dongting Luo , Wei Wang ,and Hong Zhang.

• Wenzhe Yang, Yan Zhuang, Dongting Luo , Wei Wang ,and Hong Zhang. VI-HSO: Hybrid Sparse Monocular Visual-Inertial Odometry. In IEEE Robotics and Automation Letters.

c++ Librairies (thread and system are needed). Install with

	sudo apt-get install libboost-all-dev

Linear algebra.

	sudo apt-get install libeigen3-dev

Dowload and install instructions can be found at: http://opencv.org. We tested the code on OpenCV 4.2.0.

Used for 3D visualization & the GUI. Dowload and install instructions can be found at: https://github.com/stevenlovegrove/Pangolin.

The three libraries will be compiled automatically using script build.sh.

We have tested the code on Ubuntu 20.04.

	cd VI-HSO
	chmod +x build.sh
	./build.sh

This will create the executable vi_euroc and vi_tum in bin folder.

	cd bin
	./vi_euroc ./Monocular-Inertial/EuRoC.yaml PATH_TO_SEQUENCE_FOLDER ./Monocular-Inertial/EuRoC_TimeStamps/SEQUENCE.txt FILE_NAME

	cd bin
	./vi_tum ./Monocular-Inertial/TUM_512.yaml PATH_TO_SEQUENCE_FOLDER ./Monocular-Inertial/TUM_TimeStamps/SEQUENCE.txt FILE_NAME

The result of trajectory will be saved in the bin folder.

The source code is released under GPLv3 license. We are still working on improving the code.

If you use VI-HSO in your academic work, please cite:

@ARTICLE{10218742,
author={Yang, Wenzhe and Zhuang, Yan and Luo, Dongting and Wang, Wei and Zhang, Hong},
journal={IEEE Robotics and Automation Letters}, 
title={VI-HSO: Hybrid Sparse Monocular Visual-Inertial Odometry}, 
year={2023},
volume={8},
number={10},
pages={6283-6290},
doi={10.1109/LRA.2023.3305238}}