realsense-cuda-opengl-docker

RealSense execution environment built on a Docker container on Ubuntu 20.04. NIVIDA GPU and OpenGL capable. CUADA 11.4.

1. docker build

$ git clone https://github.com/PINTO0309/realsense-cuda-opengl-docker.git \
&& cd realsense-cuda-opengl-docker \
&& docker build -t pinto0309/realsense-cuda-opengl-docker:latest .

2. docker pull

$ docker pull pinto0309/realsense-cuda-opengl-docker:latest

3. docker run

$ xhost +local: && \
docker run --gpus all -it --rm \
-v `pwd`:/home/user/workdir \
-v /tmp/.X11-unix/:/tmp/.X11-unix:rw \
--device /dev/video0:/dev/video0:mwr \
--device /dev/video1:/dev/video1:mwr \
--device /dev/video2:/dev/video2:mwr \
--device /dev/video3:/dev/video3:mwr \
--device /dev/video4:/dev/video4:mwr \
--device /dev/video5:/dev/video5:mwr \
--net=host \
-e XDG_RUNTIME_DIR=$XDG_RUNTIME_DIR \
-e DISPLAY=$DISPLAY \
--privileged \
pinto0309/realsense-cuda-opengl-docker:latest

4. Run realsense-viewer

$ realsense-viewer

5. Run nvidia-smi

$ nvidia-smi

Sat Feb  5 07:49:13 2022       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.103.01   Driver Version: 470.103.01   CUDA Version: 11.4     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  NVIDIA GeForce ...  Off  | 00000000:01:00.0  On |                  N/A |
|  0%   44C    P8    15W / 220W |    310MiB /  7959MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
+-----------------------------------------------------------------------------+

6. Run glxgears

$ glxgears

7. Running from within WSL2

[Note] OpenGL will not work properly.

1. Perform the following initial settings.
   WSL2 USB Camera Jan 17, 2022 Edition
2. Start WSL2 (Windows Terminal).

> wsl --list
Ubuntu-20.04 (default)

> wsl -d Ubuntu-20.04

3. USB attach. (Windows Terminal)

> usbipd wsl list

BUSID DEVICE                                                       STATTE
1-1   Intel(R) RealSense(TM) Depth Camera 435 with RGB Module D... Not attached

> usbipd wsl attach --busid 1-1

4. docker run. (WSL2 Terminal)

$ docker pull pinto0309/realsense-cuda-opengl-docker:latest

$ xhost +local: && \
docker run --gpus all -it --rm \
-v `pwd`:/home/user/workdir \
-v /tmp/.X11-unix/:/tmp/.X11-unix:rw \
--device /dev/video0:/dev/video0:mwr \
--device /dev/video1:/dev/video1:mwr \
--device /dev/video2:/dev/video2:mwr \
--device /dev/video3:/dev/video3:mwr \
--device /dev/video4:/dev/video4:mwr \
--device /dev/video5:/dev/video5:mwr \
--net=host \
-e XDG_RUNTIME_DIR=$XDG_RUNTIME_DIR \
-e DISPLAY=$DISPLAY \
--privileged \
pinto0309/realsense-cuda-opengl-docker:latest

$ ls -l /dev/video*
xxxxxx 1 xxxx xxxx xx,xx,xx xx:xx /dev/video0
xxxxxx 1 xxxx xxxx xx,xx,xx xx:xx /dev/video1
xxxxxx 1 xxxx xxxx xx,xx,xx xx:xx /dev/video2
xxxxxx 1 xxxx xxxx xx,xx,xx xx:xx /dev/video3
xxxxxx 1 xxxx xxxx xx,xx,xx xx:xx /dev/video4
xxxxxx 1 xxxx xxxx xx,xx,xx xx:xx /dev/video5

$ sudo chmod 777 /dev/video*

$ nvidia-smi

Sat Feb  5 07:49:13 2022       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.103.01   Driver Version: 470.103.01   CUDA Version: 11.4     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  NVIDIA GeForce ...  Off  | 00000000:01:00.0  On |                  N/A |
|  0%   44C    P8    15W / 220W |    310MiB /  7959MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
+-----------------------------------------------------------------------------+

8. CUDA assisted librealsense container

$ docker pull pinto0309/realsense-cuda-opengl-docker:latest.cuda

or

$ git clone https://github.com/PINTO0309/realsense-cuda-opengl-docker.git \
&& cd realsense-cuda-opengl-docker \
&& docker build \
-t pinto0309/realsense-cuda-opengl-docker:latest.cuda \
-f Dockerfile.cudarealsense .

$ xhost +local: && \
docker run --gpus all -it --rm \
-v `pwd`:/home/user/workdir \
-v /tmp/.X11-unix/:/tmp/.X11-unix:rw \
--device /dev/video0:/dev/video0:mwr \
--device /dev/video1:/dev/video1:mwr \
--device /dev/video2:/dev/video2:mwr \
--device /dev/video3:/dev/video3:mwr \
--device /dev/video4:/dev/video4:mwr \
--device /dev/video5:/dev/video5:mwr \
--net=host \
-e XDG_RUNTIME_DIR=$XDG_RUNTIME_DIR \
-e DISPLAY=$DISPLAY \
--privileged \
pinto0309/realsense-cuda-opengl-docker:latest.cuda