{"id":20464539,"url":"https://github.com/bytedance/schurvins","last_synced_at":"2025-04-08T11:07:40.124Z","repository":{"id":229564834,"uuid":"777049854","full_name":"bytedance/SchurVINS","owner":"bytedance","description":"[CVPR2024] SchurVINS: Schur Complement-Based Lightweight Visual Inertial Navigation System","archived":false,"fork":false,"pushed_at":"2025-01-17T05:54:06.000Z","size":23707,"stargazers_count":485,"open_issues_count":12,"forks_count":66,"subscribers_count":8,"default_branch":"master","last_synced_at":"2025-04-01T07:45:07.062Z","etag":null,"topics":["sensorfusion","slam","vins","vio"],"latest_commit_sha":null,"homepage":"","language":"C++","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/bytedance.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-03-25T05:01:25.000Z","updated_at":"2025-03-30T16:27:44.000Z","dependencies_parsed_at":"2024-03-25T06:22:47.486Z","dependency_job_id":"9f281fcd-df54-45ef-83d4-e5a12c8a3875","html_url":"https://github.com/bytedance/SchurVINS","commit_stats":null,"previous_names":["bytedance/schurvins"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bytedance%2FSchurVINS","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bytedance%2FSchurVINS/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bytedance%2FSchurVINS/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/bytedance%2FSchurVINS/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/bytedance","download_url":"https://codeload.github.com/bytedance/SchurVINS/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247829491,"owners_count":21002995,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["sensorfusion","slam","vins","vio"],"created_at":"2024-11-15T13:15:33.406Z","updated_at":"2025-04-08T11:07:40.107Z","avatar_url":"https://github.com/bytedance.png","language":"C++","funding_links":[],"categories":[],"sub_categories":[],"readme":"# SchurVINS\n## SchurVINS: Schur Complement-Based Lightweight Visual Inertial Navigation System\nWe propose a filter-based VINS framework named SchurVINS, which guarantees both high accuracy by building a complete residual model and low computational complexity with Schur complement. Technically, we first formulate the full residual model where Gradient, Hessian and observation covariance are explicitly modeled. Then Schur complement is employed to decompose the full model into ego-motion residual model and landmark residual model. Finally, Extended Kalman Filter (EKF) update is implemented in these two models with high efficiency. Experiments on EuRoC and TUM-VI datasets show that SchurVINS notably outperforms state-of-the-art (SOTA) methods in both accuracy and computational complexity. The main contributions include:\n- An equivalent residual model is proposed to deal with hyper high-dimension observations, which consists of gradient, Hessian and the corresponding observation covariance. This method is of great generality in EKF systems.\n- A lightweight EKF-based landmark solver is proposed to\nestimate position of landmarks with high efficiency.\n- A novel EKF-based VINS framework is developed to\nachieve ego-motion and landmark estimation simultaneously with high accuracy and efficiency.\n\n## 1. License\n\nThe code is licensed under GPLv3.\n\nThe SchurVINS is developed on SVO2.0(**[rpg_svo_pro_open](https://github.com/uzh-rpg/rpg_svo_pro_open)**), and thus its license is retained at the beginning of the related files.\n\n**Related Publication:**  \nYunfei Fan, Tianyu Zhao, Guidong Wang. SchurVINS: Schur Complement-Based Lightweight Visual Inertial Navigation System. (Accepted by CVPR 2024).**[PDF](https://arxiv.org/pdf/2312.01616.pdf)**.  \n\n## 2. Prerequisites\nWe have tested the codebase in **Ubuntu 18.04**.  \nThe following dependencies are needed:\n\n### System dep.\n```\n# For Ubuntu 18.04 + Melodic\nsudo apt-get install python-catkin-tools python-vcstool\nsudo apt-get install libglew-dev libopencv-dev libyaml-cpp-dev \n```\n\n\n### Ceres dep.\n\n```\nsudo apt-get install libblas-dev liblapack-dev libsuitesparse-dev\n```\n\n## 3. Build\n\nClone and build the repository:\n```\ncd ~/catkin_ws/src\ngit clone https://github.com/bytedance/SchurVINS.git\nsource ~/catkin_ws/devel/setup.bash\nmkdir -p ~/catkin_ws/src/SchurVINS/results\nmkdir -p ~/catkin_ws/src/SchurVINS/logs\nvcs-import \u003c ./SchurVINS/dependencies.yaml\ntouch minkindr/minkindr_python/CATKIN_IGNORE\ncatkin build\n```\n\n## 4. Run\nWe provide examples to run SchurVINS with [EuRoC dataset](https://projects.asl.ethz.ch/datasets/doku.php?id=kmavvisualinertialdatasets#downloads). \n\nOpen three terminals, launch the schur_vins , rviz and play the bag file respectively. Take MH_01 for example \n\n```\nroslaunch svo_ros euroc_vio_stereo.launch\nrviz -d svo_ros/rviz_config.rviz\nrosbag play YOUR_PATH_TO_DATASET/MH_01_easy.bag \n```\n# 5. Reproducibility\nThere might be minor differences between the released version and the results in the paper. Please note that multi-thread performance has some randomness due to CPU utilization.\nOn EuRoC datasets, please note to skip the first few seconds of datasets with the following configurations for reproducibility.\n\n\n|||||||||||\n|----|------|----|----|------|----|----|------|----|----|\n|**MH1**|**MH2**|**MH3**|**MH4**|**MH5**|**V11**|**V12**|**V13**|**V21**|**V22**|\n|30s|30s|10s|13s|15s|0s|0s|2s|0s|0s|\n|||||||||||\n\n# 6. Security\n\nIf you discover a potential security issue in this project, or think you may\nhave discovered a security issue, we ask that you notify Bytedance Security via our [security center](https://security.bytedance.com/src) or [vulnerability reporting email](sec@bytedance.com).\n\n# 7. Acknowledgement\nThis work incorporates the well-known SVO2.0 open-source code. We extend our gratitude to the authors of the software.\n- [rpg_svo_pro_open](https://github.com/uzh-rpg/rpg_svo_pro_open)\n\n\n# 8. Citation\nIf you found this code/work to be useful in your own research, please considering citing the following information. Additionally, please considering citing SVO2.0([rpg_svo_pro_open](https://github.com/uzh-rpg/rpg_svo_pro_open)) since SchurVINS is developed on it.\n```bibtex\n@inproceedings{fan2024schurvins,\n  title={SchurVINS: Schur Complement-Based Lightweight Visual Inertial Navigation System},\n  author={Fan, Yunfei and Zhao, Tianyu and Wang, Guidong},\n  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},\n  pages={17964--17973},\n  year={2024}\n}\n```\n\n# 9. We are Hiring!\nOur team is hiring FTEs with background in Deep Learning, SLAM, and 3D Vision. We are based in Beijing and Shanghai. If you are interested, please send your resume to frank.01[AT]bytedance[DOT]com.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbytedance%2Fschurvins","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fbytedance%2Fschurvins","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fbytedance%2Fschurvins/lists"}