Skip to content
93wonPublic

About

No description, website, or topics provided.

Resources

Stars

403 stars

Watchers

7 watching

Forks

Repository files navigation

OpenVIO License: MIT

OpenVIO is an MIT-licensed visual-inertial odometry system for researchers and developers.

Paper | Citation

OpenVIO demo

Supported configurations

Camera model Monocular Stereo Tested on
360° equirectangular, two-lens stitched (EQUIRECTANGULAR) ✓ – PanoAir
Pinhole, radial-tangential (PINHOLE_RADTAN) ✓ ✓ EuRoC MAV
Fisheye, equidistant (FISHEYE_EQUIDISTANT) ✓ ✓ TUM-VI

Online calibration

Pinhole Fisheye 360°
Camera-IMU time offset ✓ ✓ ✓
Camera-IMU extrinsics ✓ ✓ ✓
Time offset between the two lenses – – ✓
Stitching error – – ✓

Features

  • Mono and stereo with one estimator; a new lens model is one file
  • Initialization at rest or in motion
  • Tracking under auto-exposure changes and night lights
  • Lost landmarks found again after occlusion and image-border crossing
  • Deterministic: the same input gives the same trajectory
  • Optional ROS 1 and ROS 2 nodes with separate build_ros1.sh and build_ros2.sh scripts

Benchmark

Benchmark

Mean over the completed sequences. Details in the release notes.

Docs

Quick start doc/usage.md
ROS 1 / ROS 2 doc/ros.md
Release notes v1.0.1
Datasets doc/datasets.md
Configuration doc/configuration.md
Camera models doc/camera_models.md
Architecture doc/architecture.md
References doc/references.md

Citation

If you find OpenVIO useful in your research, please cite our paper:

@article{choi2026openvio,
  title   = {{OpenVIO}: A Versatile Open-Source Visual-Inertial Odometry System},
  author  = {Choi, Seungwon},
  journal = {arXiv preprint arXiv:2610.41671},
  year    = {2026},
  url     = {https://arxiv.org/abs/2610.41671}
}

Benchmark Datasets

@article{wu2026panoair,
  title   = {PanoAir: A Panoramic Visual-Inertial SLAM with Cross-Time Real-World UAV Dataset},
  author  = {Wu, Yiyang and Zhang, Xiaohu and Du, Yanjin and Zhang, Tongsu and Li, Chujun and Chen, Siyang and
             Zhang, Guoyi and Xu, Xiangpeng},
  journal = {arXiv preprint arXiv:2604.00852},
  year    = {2026}
}

@article{burri2016euroc,
  title   = {The EuRoC micro aerial vehicle datasets},
  author  = {Burri, Michael and Nikolic, Janosch and Gohl, Pascal and Schneider, Thomas and Rehder, Joern and
             Omari, Sammy and Achtelik, Markus W and Siegwart, Roland},
  journal = {The International Journal of Robotics Research},
  volume  = {35},
  number  = {10},
  pages   = {1157--1163},
  year    = {2016}
}

@inproceedings{schubert2018tumvi,
  title     = {The TUM VI Benchmark for Evaluating Visual-Inertial Odometry},
  author    = {Schubert, David and Goll, Thore and Demmel, Nikolaus and Usenko, Vladyslav and St{\"u}ckler, J{\"o}rg and
               Cremers, Daniel},
  booktitle = {2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  pages     = {1680--1687},
  year      = {2018}
}

About

No description, website, or topics provided.

Resources

Stars

403 stars

Watchers

7 watching

Forks

Releases

Packages

Contributors

Languages