Skip to content

Latest commit

 

History

History
152 lines (99 loc) · 9.13 KB

dockerhub.md

File metadata and controls

152 lines (99 loc) · 9.13 KB
Raw

ZED SDK Docker Images

These images let you use the ZED SDK with docker, even with the ZED camera connected (or an SVO file). https://github.com/stereolabs/zed-docker

Base Image Tags

All images are based on Nvidia CUDA. You need nvidia-docker to run them, see below. Please note that these images can't run graphical programs, including the tools. It requires openGL support (see gl-devel tags). All desktop images now include the ZED Python API.

  • runtime images are the lightest and comes with every dependency installed. It's only meant to run applications linked with the ZED SDK.

  • devel images contains all development tools to compile application including the cuda toolchain, static libraries, and headers of CUDA and the ZED SDK.

  • gl-devel images include openGL support to be able to run the tools and sample. It also contains the development tools similarly to devel image. Please note that some images variants are not available since Nvidia didn't release image for recent version of CUDA and Ubuntu https://gitlab.com/nvidia/container-images/cuda/-/issues/147

Jetson specific tags

This the jetsons are usually much more limited in storage, more variants are available to limit the disk space usage. Unlike the desktop images, the OpenGL support is included from the base image provided by Nvidia, and therefore available on all images.

  • py-devel images contains all development tools to develop applications that uses ZED Python API.

  • py-runtime images are the lightest and comes with every dependency installed. It's only meant to run applications that uses the ZED Python API.

  • tools-devel images include the tools and sample and the development tools similarly to devel image. It is the most complete and biggest image.

LATEST ZED SDK 4.0 tags

Ubuntu 22.04

CUDA 12.1

CUDA 11.8

Ubuntu 20.04

CUDA 12.1

CUDA 11.8

Ubuntu 18.04 (deprecated)

CUDA 11.8

CUDA 10.2

Jetson L4T 35 (JetPack 5)

L4T 35.3 (JP 5.1.1)

L4T 35.2 (JP 5.1.0)

L4T 35.1 (JP 5.0.2)

Jetson L4T 32 (JetPack 4)

L4T 32.7 (JP 4.6.X)

Running Containers

docker run --gpus all -it --privileged stereolabs/zed:4.0-runtime-cuda11.8-ubuntu18.04

--privileged option is used to pass through all the device to the docker container, it might not be very safe but provides an easy solution to connect the USB3 camera to the container.

Display support

To run it, we need to add the right to connect to the X server : ​

xhost +si:localuser:root

​ Then to run it : ​

docker run --gpus all -it --privileged -e DISPLAY -v /tmp/.X11-unix:/tmp/.X11-unix stereolabs/zed:4.0-gl-devel-cuda11.4-ubuntu18.04

​ Any OpenGL tools are now able to run, for instance :

/usr/local/zed/tools/ZED_Explorer

Notes

Docker Setup

Since we need CUDA, NVIDIA Container Toolkit must be used (except for compilation only).

Follow the instructions at https://github.com/NVIDIA/nvidia-docker

Using docker on Tegra

With the recently added support of nvidia docker, it is now possible to run the ZED SDK inside docker on Jetson. We now provide a compatible image :

docker pull stereolabs/zed:4.0-devel-l4t-r35.1

The image is based on the NVIDIA L4T image.

More information

Check out the README on Github