#175 · Primary category: Computer Vision

zed-sdk

3d-reconstruction computer-vision depth-camera depth-estimation human-pose-estimation machine-learning object-detection perception point-cloud realtime slam stereo stereo-vision stereolabs zed zed-camera zed-x zed2

⚡️The spatial perception framework for rapidly building smart robots and spaces

Project last updated:06/18/26

GitHub Stars

1.2K

Forks

507

Contributors

19

License

MIT

Why we included this project

If you are putting a depth or stereo camera into a real product, whether a mobile robot, an autonomous vehicle testbed, or a factory inspection rig, this repo is a practical place to start. It is the official companion to Stereolabs' ZED cameras and ships the SDK as runnable tutorials and samples, so you can move from your first image grab to depth maps, 3D point clouds, SLAM, and object and human pose detection in a few guided steps. The same perception pipeline runs on desktop GPUs and embedded Jetson boards, which lets you develop locally and then move to the edge. The newer custom depth mode even accepts disparity from your own stereo network and runs the rest of the pipeline on it. The samples cover day-to-day production needs like recording and multi-camera setups, so treat this as SDK glue and reference code to pair with your own application logic rather than a standalone tool.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category