Skip to content
RobothecaBeta
Log in

OpenVINS

  • Software
  • Navigation and mapping
  • Library

OpenVINS is an open-source visual-inertial state estimation system built on a filter-based MSCKF design, intended as a research platform for visual-inertial estimation.

Made by Robot Perception & Navigation Group (RPNG), University of Delaware, United States (checked on github.com/rpng (opens in a new tab)).

Source on GitHub: github.com/rpng/open_vins (opens in a new tab). Facts read at commit 6948812 (opens in a new tab).

Licenses

Code
GPL-3.0-only (opens in a new tab)

Read by Robotheca from LICENSE (opens in a new tab) (GPL-3.0-only) at commit 6948812 (opens in a new tab), on 16 Oct 2026.

Commercial use
Yes

Copyleft: Modified source must be published.

What it does

  • Core is an Extended Kalman filter that fuses inertial data with sparse visual feature tracks using a sliding-window MSCKF formulation.
  • Supports five feature representations, including global and anchored XYZ and inverse-depth forms.
  • Calibrates camera-to-IMU transform, time offset, camera intrinsics, and inertial intrinsics.
  • Supports monocular, stereo, and binocular cameras with KLT or descriptor-based tracking and masked tracking.
  • Includes static and dynamic initialization, zero velocity detection and updates, and ARUCO tag or sparse feature SLAM features.
  • Provides out-of-the-box evaluation on EuRoC MAV, TUM-VI, UZH-FPV, KAIST Urban and other VIO datasets, with an evaluation suite (ATE, RPE, NEES, RMSE).

Written by Robotheca from the project's README (opens in a new tab) at commit 6948812.

Getting started

Difficulty
Advanced

Maintenance

Status
Maintained: a commit in the last 12 months
Newest commit
30 Nov 2025 on master
Latest release
Develop v2.7 (20 Jun 2023)

Checked on GitHub on 16 Oct 2026.

Links

Robots on Robotheca that use it

No robot on Robotheca is linked to it yet. Links come from a robot's robot.yaml, from its dependency files read as text, or from Robotheca's editors.

Robotheca lists software as a reference: nothing is hosted or run here, and every fact links to where it was read.