Closed firefly2442 closed 2 years ago
This is something between ffmpeg and the camera. It should be reproducible when running just ffmpeg from host machine (without watsor). Is the error the same if only one camera is connected?
With regard to cuvid I remember that I could activate it by using only the following option and no -hwaccel
:
ffmpeg -hide_banner -c:v h264_cuvid -i sample.mp4 -f null /dev/null
Thanks, this seems to work:
ffmpeg:
decoder:
- -hide_banner # hide build options and library versions
- -loglevel
- error
- -nostdin
#- -hwaccel # These options enable hardware acceleration, check what's available with: ffmpeg -hwaccels
#- cuvid
#- -hwaccel_output_format
#- yuv420p
- -c:v
- h264_cuvid # use GPU for h264 decoding
- -fflags
- nobuffer
- -flags
- low_delay
- -fflags
- +genpts+discardcorrupt
- -i # camera input field will follow '-i' ffmpeg argument automatically
- -f
- rawvideo
- -pix_fmt
- rgb24
#- -rtsp_transport # try to prevent lost packets/frames via TCP
#- tcp
Hello. I have three cameras all running h264 streams via RTSP. I have an NVIDIA 970 GPU. I'm looking to use the GPU for decoding the h264 stream as well as performing the object detection. I am seeing some decoding errors in the logs.
Shelling into the container I see
cuvid
as an option.nvidia-smi
andnvtop
in the main OS show watsor leveraging the GPU.The
health
endpoint shows the following:My guess is it has something to do with what I'm passing as parameters to ffmpeg. Here is my
config.yaml
: