FFmpeg Camera
The ffmpeg
platform allows you to use every video feed with FFmpeg as camera in Home Assistant. The input for ffmpeg need to support that could have multiple connection to source (input) in same time. For every user in UI and all 10 seconds (snapshot image) it make a new connection/reading to source. Normally that should never be a trouble only in strange selfmade constructs can be make mistakes.
You need a ffmpeg
binary in your system path. On Debain 8 you can install it from backports. If you want Hardware support on a Raspberry Pi you need tobuild from source by yourself. Windows binary are avilable on the FFmpeg website.
To enable your FFmpeg feed in your installation, add the following to your configuration.yaml
file:
# Example configuration.yaml entry
camera:
- platform: ffmpeg
input: FFMPEG_SUPPORTED_INPUT
name: FFmpeg
ffmpeg_bin: /usr/bin/ffmpeg
extra_arguments: -q:v 2
Configuration variables:
- input (Required): A ffmpeg compatible input file, stream or feed.
- name (Optional): This parameter allows you to override the name of your camera.
- ffmpeg_bin (Optional): Default ‘ffmpeg’.
- extra_arguments (Optional): Extra option they will pass to
ffmpeg
. i.e. image quality or video filter options.
Image quality
You can control the image quality
with extra_arguments
-q:v 2-32
or with lossless option -pred 1
.
Troubleshooting
In most of case, ffmpeg
autodetect all needed options to read a video/audio stream or file. But it is possible in rare cases that’s needed to set a option to help ffmpeg
. Per default ffmpeg
use 5 seconds to detect all options or abort.
First check, if your stream playable by ffmpeg
with (use option -an
or -vn
to disable video or audio stream):
$ ffmpeg -i INPUT -an -f null -
Now you can see what going wrong. Following list could be help to solve your trouble:
[rtsp @ ...] UDP timeout, retrying with TCP
: You need to set RTSP transport in the configuration with:input: -rtsp_transport tcp -i INPUT
[rtsp @ ...] Could not find codec parameters for stream 0 (Video: ..., none): unspecified size
: FFmpeg need more data or time for autodetect. You can set theanalyzeduration
and/orprobesize
option, play with this value. If you know the needed value you can set it with:input: -analyzeduration xy -probesize xy tcp -i INPUT
. More information about that can be found on FFmpeg.