while the current stable version works for me, the beta (3.0.0) stops saying "CUDA driver version is insufficient".
I don't have a nvida card.
I launch the app on debian bullseye using "snap run fakecam" as simply typing "fakecam" results in "command not found".
Both versions, stable and beta never give me a GUI. (I'm using xfce4 on lightdm).
WARNING: cgroup v2 is not fully supported yet, proceeding with partial confinement
No background specified. will blur your background instead.
Starting capture using device: /dev/video0
Resolution: 640:480
Process Process-1:
Traceback (most recent call last):
File "/snap/fakecam/111/usr/lib/python3.6/multiprocessing/process.py", line 258, in _bootstrap
self.run()
File "/snap/fakecam/111/usr/lib/python3.6/multiprocessing/process.py", line 93, in run
self._target(*self._args, **self._kwargs)
File "/snap/fakecam/111/lib/python3.6/site-packages/fakecam/capture.py", line 153, in start
frame = get_frame(cap, scaler, ones, dilation, background=background_scaled, use_hologram=use_hologram, height=height, width=width)
File "/snap/fakecam/111/lib/python3.6/site-packages/fakecam/capture.py", line 81, in get_frame
mask = get_mask(frame, scaler, dilation, height=height, width=width)
File "/snap/fakecam/111/lib/python3.6/site-packages/fakecam/capture.py", line 33, in get_mask
results = cvNet.forward()[0][0]
cv2.error: OpenCV(4.5.1) /root/parts/opencv/src/modules/dnn/src/cuda4dnn/csl/memory.hpp:54: error: (-217:Gpu API call) CUDA driver version is insufficient for CUDA runtime version in function 'ManagedPtr'
Hello,
while the current stable version works for me, the beta (3.0.0) stops saying "CUDA driver version is insufficient".
I don't have a nvida card.
I launch the app on debian bullseye using "snap run fakecam" as simply typing "fakecam" results in "command not found".
Both versions, stable and beta never give me a GUI. (I'm using xfce4 on lightdm).
How can I get a UI and turn off usage of CUDA?