Docker Images with Demos and Products Evaluation Versions: Docker image installation

Requesting Docker images

These images are available in the docker hub of RidgeRun. These images are meant to be run by Jetson Platforms. To get access you will need the following dependencies.

Dependencies

Docker

In your target board, you will need to have docker installed.

NVIDIA Runtime

This runtime can be installed by following the next guide from [| NVIDIA]

Download the image

In the next table, you can find available images by their tag name for different JetPacks versions and if they include or not deepstream.

Table 1. Features provided in the evaluation version
TAG	JetPack Version
jetson_4x_v1.0	4.x
jetson_5x_v1.0	5.x

From this table, you can pull your desired image as follows. In this tutorial, IMAGE will represent your selected image.

export IMAGE=ridgerun/products-evals-demos:<IMAGE_CHOOSEN>

docker pull $IMAGE

Once you have download the image, now you can verify it as follows.

docker images

And you will see in the output the name of the image.

REPOSITORY                       TAG              IMAGE ID       CREATED          SIZE
ridgerun/products-evals-demos   jetson_4x_v0.1   7832cc613a31   53 seconds ago   10.6GB
ridgerun/products-evals-demos   jetson_5x_v0.1   34cd5f9d1d82   2 weeks ago     13.8GB

JP 5.1.2

This is a provisional image that only works for JP 5.1.2, you can download it as follows.

IMAGE=ridgerun/products-evals-demos:jetson_5.1.2_v0.3
docker pull $IMAGE

The following steps are the same as the other versions, but take into account that the version default of DeepStream is 6.3 instead of 6.2.

Building container

Base image

For this step you can modify it regarding your preferences, but for the basic usage of the image you can run the following.

docker run --name demo_rr --net=host --runtime nvidia --ipc=host -v /tmp/.X11-unix/:/tmp/.X11-unix/ -w /RR_EVALS  -v /tmp/argus_socket:/tmp/argus_socket --cap-add SYS_PTRACE --privileged --volume="$HOME/.Xauthority:/root/.Xauthority:rw" -v /dev:/dev -v /proc:/writable_proc -e DISPLAY=$DISPLAY -it $IMAGE /RR_EVALS/demo.sh

Now it will open the docker container in eval mode where you can test the RidgeRun products available.

gst-inspect rtspsink

Whenever you need to run the image again, please use the following commands.

docker start demo_rr

docker attach demo_rr

With Deepstream

Before you run the build command with deepstream, make sure you have deepstream installed in your board. You can check it with

ls /opt/nvidia/deepstream/

Now create the next variables with the next path. Take for example deesptream 6.2:

DEEP_PATH=/opt/nvidia/deepstream/deepstream-6.2

DEEP_LIB_PATH=/opt/nvidia/deepstream/deepstream-6.2/lib/libnvdsgst_customhelper.so

Now run the next command to create the container using the downloaded image.

docker run --name demo --net=host --runtime nvidia --ipc=host -v /tmp/.X11-unix/:/tmp/.X11-unix/ -v /tmp/argus_socket:/tmp/argus_socket --cap-add SYS_PTRACE --privileged --volume="$HOME/.Xauthority:/root/.Xauthority:rw" -v /dev:/dev -v /proc:/writable_proc --volume $DEEP_PATH:$DEEP_PATH -w /RR_EVALS -v  /usr/lib/aarch64-linux-gnu/gstreamer-1.0/deepstream/:/usr/lib/aarch64-linux-gnu/gstreamer-1.0/deepstream/ -v $DEEP_LIB_PATH:$DEEP_LIB_PATH -e DISPLAY=$DISPLAY -e LD_LIBRARY_PATH=$DEEP_PATH/lib -it $IMAGE /RR_EVALS/demo.sh

This will open the container in evaluation mode.

Index

Next: Products Evaluation

❯