For AI agents: a documentation index is available at the root level at /llms.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
The ZED SDK provides multiple code examples to facilitate your learning experience with spatial perception and AI. All of these samples can be found on the Stereolabs GitHub page.
If you need assistance with building the examples, you can refer to the API Language section of the documentation for detailed instructions on how to set up your development environment.
If you haven’t already, we recommend checking out our tutorials first. These resources can provide valuable information to help you get started with the ZED SDK.
Stream ZED video over the network — single or multiple senders, a receiver, encoded passthrough, and H.264 record-while-send — for remote viewing or processing.
Extract and visualize depth maps and point clouds, with variants for multi-camera capture, region of interest, image refocus, custom depth ingestion, and fusion.
Detect and track objects in 3D — bird’s-eye and image viewers, concurrent detection with body tracking, custom detectors (YOLO/Ultralytics), and multi-camera fusion.
Detect and track 3D human body skeletons, export the data to FBX/JSON, integrate with Unity/Unreal Engine, or fuse multiple cameras for whole-space tracking.
Record and play back SVO files — single or multi-camera, with external/sensor data — and export SVO content to common video, image and sensor file formats.
Fuse a real GNSS receiver with the camera’s visual-inertial odometry for global-scale localization, with live, recording, playback and map-server samples.
Manage multiple ZED cameras, ZED X One cameras and LiDARs through a single unified API, live or from recordings, with an interactive multi-sensor placement/calibration tool.
