Robot/Camera Calibration

Extrinsic and Intrinsic Camera and Camera-to-Robot Calibration

We use the right camera model, calibration pattern and algorithms for your specific use case.

All vision technology depends on properly calibrated cameras. We create static or dynamic calibration procedures, calibrating intrinsic camera parameters, or estimating at run-time the extrinsic camera parameters.

We provide:

  • Selection of Camera Calibration Patterns
  • Camera model selection and implementation
  • Online estimation of extrinsic properties of camera with respect to other cameras, imu, wheels
  • Unsynchronized, non-overlapping field of view multi-camera extrinsic calibration
  • Camera to wheel base (odometry) calibration
  • Tuning and improving calibration algorithms
  • Multi-camera extrinsic calibration

We can calibrate any 2D or 3D camera, such as the Stereolabs ZED, Lucid Triton HDR, Intel Realsense and Luxonis OAK.

Open Source Libraries

We frequently use

and have our own calibration stack which overcomes well known limitations of the above libraries.

Learn more from our blog

ZED X and NVIDIA: Applications, Architecture and Integrations

With their ZED X camera, Stereolabs has chosen for a firm integration with the Nvidia Jetson platform. Here’s a high level overview of what applications are supported, and also some pointers to how the ZED+Nvidia architecture works!

Read ZED X and NVIDIA: Applications, Architecture and Integrations
A Quadruped being diagnosed on a table

Real-Time Robotics: The Extended Kalman Filter

In this last part, we’ll explain how an Extended Kalman Filter works, and which libraries allow you to implement one in robotics.

Read Real-Time Robotics: The Extended Kalman Filter

Real-Time Robotics: How a Kalman Filter Can (not) Help

In this Part 2 of our mini series, we'll explore the Kalman Filter and if it can be used to enhance sensor quality and build real-time robotics applications.

Read Real-Time Robotics: How a Kalman Filter Can (not) Help