OS
Ubuntu 20.04.4 LTS 64-bit
Graphics
Mesa Intel® Xe Graphics (TGL GT2)
Steps followed / commands executed
I followed steps from readme. Runnning v4l2-ctl --list-devices
shows:
$ v4l2-ctl --list-devices
Dummy video device (0x0000) (platform:v4l2loopback-000):
/dev/video20
Integrated Camera: Integrated C (usb-0000:00:14.0-8):
/dev/video0
/dev/video1
I updated the cameraID
setting in ./config.yaml
to just use the number as so:
# change input camera device ID (just add the ID number alone)
cameraID = 0
Running python3 main.py
shows this output (with the Processing frame
logs continuing until I stop the script):
2022-07-08 08:38:07.014005: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/manny/.local/lib/python3.8/site-packages/cv2/../../lib64:/usr/local/lib
2022-07-08 08:38:07.014018: I tensorflow/stream_executor/cuda/cudart_stub.cc:29] Ignore above cudart dlerror if you do not have a GPU set up on your machine.
set format result (0 is good):0
INFO: Created TensorFlow Lite XNNPACK delegate for CPU.
2022-07-08 08:38:08.138452: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcuda.so.1'; dlerror: libcuda.so.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /home/manny/.local/lib/python3.8/site-packages/cv2/../../lib64:/usr/local/lib
2022-07-08 08:38:08.138474: W tensorflow/stream_executor/cuda/cuda_driver.cc:269] failed call to cuInit: UNKNOWN ERROR (303)
2022-07-08 08:38:08.138486: I tensorflow/stream_executor/cuda/cuda_diagnostics.cc:156] kernel driver does not appear to be running on this host (manny-ThinkPad-E15-Gen-2): /proc/driver/nvidia/version does not exist
2022-07-08 08:38:08.138705: I tensorflow/core/platform/cpu_feature_guard.cc:193] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 AVX512F AVX512_VNNI FMA
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Processing frame: 0 at 21 fps. Press ctrl+q to stop
Processing frame: 1 at 33 fps. Press ctrl+q to stop
Processing frame: 2 at 32 fps. Press ctrl+q to stop
Processing frame: 3 at 30 fps. Press ctrl+q to stop
...
I then fire up a camera app (e.g., MS Teams) but the camera only displays a black image, no video and no virtual or blurred background.
![teams_virtual_background_sad](https://user-images.githubusercontent.com/13630322/178031875-7ed148ee-4226-4cde-85fc-de5b3d4dba4a.png)
If I have the physical camera in use before running the python3 main.py
script, then I get the error:
[ WARN:[email protected]] global /io/opencv/modules/videoio/src/cap_v4l.cpp (889) open VIDEOIO(V4L2:/dev/video0): can't open camera by index
So I assume it is correct to run the script prior to starting the physical camera.
Am I missing a step or anything?
Edit: After doing some more digging, I realize perhaps this is due to my graphics card not being supported sufficiently for running TensorFlow? If that's the case, feel free to close this issue.