Recherche avancée

Sur d’autres sites (6312)

• Concatenate multiple videos with a black screen loop into one video

  21 mars 2016, par AHC

  I am using ffmpeg to join a bunch of videos together.
  I am using the classic join ffmpeg code :
  

  ffmpeg -f concat -i joinlist.txt -c copy joinedfile.mp4

  but the problem is that the videos are of different formats, encodings, but the same size : all 640x480. I want to join them all and put a black screen video with no sound every other video :

  video1 + black_screen_video + video2 + black_screen_video + video3 ...

  I generated a black screen video of 2 seconds duration using :

  ffmpeg -f lavfi -i color=c=black:s=640x480:d=2 black_screen_video.mp4

  so all of the videos are of the same size : 640x480, and 25 fps but different codecs. The videos have sound, except for the black screen video.
  I can’t do anything manual, because the number of videos are around several hundred. So it has got to be an automatic way to do all this.

  When I joined them together using the above code, the resulting video does not play correctly at all.

  I know that I have to re-encode them, but how to do this to all these videos at once, with one line of code ?

  Update :
  I am already using with success this code to join them together, but only three, if I have more than one hundred, it is time consuming to write down one by one :

  ffmpeg -i vid1.avi -i vid2.avi -i vid3.avi -filter_complex "[0:v:0] [0:a:0] [1:v:0] [1:a:0] [2:v:0] [2:a:0] concat=n=3:v=1:a=1 [v] [a]" -map "[v]" -map "[a]" output.mp4

  but this is joining only the videos, not looping the black screen video. When I do with black screen, ffmpeg gives me stream matching errors.

  update :

  here is the console image :
  

  2nd update :

  a very long list of errors in red, of which a screenshot here :
  

• Capturing and processing a live RTMP stream

  8 février 2018, par Yantor

  I’m trying to download a live stream (not a file) coming from a live camera feed available at the following website : http://www.dot.ca.gov/video/.

  I used Wireshark for sniffing the TCP packets and was able to extract the RTMP parameters, but wasn’t able to use them with FFMPEG/VLC for downloading / playing the stream on VLC (I guess I didn’t construct the URL correctly).

  for example, for the camera feed available here, I got the following parameters :

  swfUrl : http://www.dot.ca.gov/research/its/StrobeMediaPlayback.swf

  pageUrl : http://www.dot.ca.gov/d4/d4cameras/ct-cam-pop- N17_at_Saratoga_Rd.html

  tcUrl : rtmp ://wzmedia.dot.ca.gov:1935/D4
  Play : E37_at_Lakeville_Rd.stream.

  Is there a chance someone is familiar with this and can help with understanding how I can use the above for downloading the stream ?

  Thanks a lot ! Yaniv

• ffmpeg : How do you skip frames in an image sequence input ?

  15 mai 2021, par MirceaKitsune

  I'm rendering an animation in Blender as a png image sequence, which I then compile to a mp4 video with ffmpeg. The animation is normally 60 FPS but for testing I reduce it to 30 FPS : I do this by enabling frame skipping at render time. My issue is getting ffmpeg to understanding how to pick images every two numbers rather than one.

  


  ffmpeg -start_number 1 -framerate 60 -f image2 -i ./video/%04d.png -b:v 32000k ./render.mp4


  


  This is what I normally use and it works great outside of this circumstance. Obviously I set -framerate 30 in this case. Problem is the command still expects the following image order :

  


  /video/0001.png
/video/0002.png
/video/0003.png
/video/0004.png
/video/0005.png
/video/0006.png
/video/0007.png
/video/0008.png


  


  Whereas the pattern it needs to work with during testing is :

  


  /video/0001.png
/video/0003.png
/video/0005.png
/video/0007.png


  


  I saw a suggestion to add -vf "tblend=addition,framestep=2" but both this or not changing the command at all produces a 0 byte (or very tiny) mp4 file, likely due to only including the first frame and dropping the rest.