Recherche avancée

Sur d’autres sites (6508)

• Creating personalized video from user submitted pictures and data

  18 décembre 2016, par Rajat Singhal

  Can someone guide me towards what technology to use to create personalized video from user submitted pictures and data.. The process has to be automated, as in server must be able to create a downloadable video from photos and text submitted by user..

  So the process will probably be like one video will be created by a hired artist, with placeholders where the user submitted pictures and text will fit in. Now with user submitted data video can be created and downloaded from the website..

  An example can be the videos created by facebook now-a-days on your birthday or year end. They consist some of your photos, some text and have a common video theme.. You can view one here http://newsroom.fb.com/news/2016/12/facebook-2016-year-in-review/

  One way I’ve found is to write the video code in html5 and then record it with phantomjs and ffmpeg.. http://mindthecode.com/recording-a-website-with-phantomjs-and-ffmpeg/

  But it seems a bit unnatural way of doing it.. And also I think not a lot of good artists are out there who can create the video theme in html5..

• Detecting on which frame there is audio presence on a video

  26 août 2016, par drov

  I have a few video files that corresponds to a tv zapping (one channel with sound, then a black screen without sound, then sound again with the new channel)

  I already detect pretty much everything but I would like to know how long it takes for the audio to appear after the end of the black screen.

  Basically I extract the audio from the video and giving the starting frame I would like to know at which frame there is some audio again.

  Then using that I can easily calculate the time it took for the audio to appear.

• ffmpeg - Drawing rotated text on video with complex filters takes a very long time

  14 mai 2019, par Bedrule Paul

  I am trying to overlap different text pieces on some placeholders in a video, and I am using multiple complex filters of the following type :

  ffmpeg -i ~/Desktop/input.mp4 -filter_complex  \
  
      "color=black@0:100x100,format=yuva444p[c]; \
  
      [c][0]scale2ref[ct][mv31]; \
  
      [ct]setsar=1,split=1[t31];\
  
      [t31]\
  
      drawtext=text='text':x='main_w/2-text_w/2+70':y=210:fontsize="100":fontcolor=black,\
  
      drawtext=text='text2':x='main_w/2-text_w/2+75':y=340:fontsize="100":fontcolor=black,\
  
      rotate=-0.07:ow=rotw(-0.07):oh=roth(-0.07):c=black@0[txta31]; \
  
      [mv31][txta31]overlay=enable='between(t, 0, 1.15)':x='min(0,-H*sin(-0.07))':y='min(0,W*sin(-0.07))':shortest=1" \
  
         ~/Desktop/result.mp4 -y1

  My goal is to write differently rotated texts on different time intervals in the video. The problem is that at about 10-12 [t31]-like pieces(here is an example of only one command), the rendering time of the video is twice the time of the video, whereas drawing straight horizontal text takes about 10-20% of the total video length (examples, for a 1 minute video, it takes about 8-10 seconds to write straight horizontal text, and about 2 minutes to write the same amount of text, but inclined with an angle). Is there any better way to do these multiple rotated text bits with more performance ?