The VideoLAN Forums

Hello,

I'm running vlc 1.1.12 on Ubuntu 11.10.
Using webcam as input,I'm trying to run vlc as an RTSP server, and in the same time I'm trying make vlc save snapshots of the video that it captures.
I have no problem with the transcoding part. It's that I failed to make vlc save snapshots while disabling the output display (even with using -Idummy flag).

For simplicity, I've removed the code I'm using for transcoding. The following is the code I'm using for saving snapshots: (Remove the --scene-replace flag if you want each captured jpeg saved in a new flie).
Note that although the -Idummy flag is used, however a display window still appears (even even if cvlc instead of vlc).

I've tried using cvlc on terminal and redirecting the standard output to > /dev/null, however since the fact that output is generated, so this uses a relatively lot of processing power (my vlc cmd will be deployed on a node with small processing power).

So, the question is:
How to save snapshots while completely disabling output window??.

You can turn off video output with -Vdummy, but VLC will still consume a lot of CPU power to decode the video anyway.

It works. Thanks
It is still mystery to me that if I only transcoded the video then the processing power is still relatively low. However, to take snapshots: I have to duplicate the sout-stream and send one stream to be rtsp server and send the other to display (so that the scene video-filter can be applied to the display stream). If I tried to apply the scene video-filter without a display stream then the command doesn't work (i.e. it doesn't take snapshots).