Skip to content
RobothecaBeta
Log in

OKVIS2

  • Software
  • Navigation and mapping
  • Library

OKVIS2 is a visual-inertial SLAM system that estimates a device's pose and builds a map from camera and IMU data, using keyframes and nonlinear optimisation with loop closure.

Made by Mobile Robotics Lab.

Source on GitHub: github.com/ethz-mrl/okvis2 (opens in a new tab). Facts read at commit a2ea006 (opens in a new tab).

Licenses

Code
BSD-3-Clause (opens in a new tab)

Read by Robotheca from LICENSE (opens in a new tab) (BSD-3-Clause) at commit a2ea006 (opens in a new tab), on 16 Oct 2026.

Commercial use
Yes

What it does

  • Estimates trajectory and state (pose, velocity, gyro and accelerometer biases) from stereo camera and IMU measurements.
  • Uses keyframe-based nonlinear optimisation and supports loop closure for scalable, realtime SLAM.
  • Provides a library where client code maintains a trajectory object via optimisation and IMU callbacks.
  • Includes a demo application that processes EuRoC-format datasets and supports Realsense D435i and D455 cameras.
  • Can run as ROS2 nodes, processing live Realsense input, subscribed topics, recorded bags or datasets.
  • Optionally runs a LibTorch segmentation network to remove points such as sky, which can use an NVIDIA GPU.

Written by Robotheca from the project's README (opens in a new tab) at commit a2ea006.

Getting started

Difficulty
Advanced
Hardware it needs, as the project states it
It describes setup for Ubuntu and OSX (Intel and M1).

Maintenance

Status
Maintained: a commit in the last 12 months
Newest commit
11 Dec 2025 on main
Latest release
None published on GitHub

Checked on GitHub on 16 Oct 2026.

Links

Robots that can run 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.