Recherche avancée

Sur d’autres sites (8481)

• ffmpeg do stream copy together with draw text

  8 mars 2016, par Haris

  I need to record an RTSP stream, before storing it I need draw some text on the video.

  And the below command works fine for recording stream only(no text write).

  openRTSP -D 10 -v -t -c -b 800000 rtsp://video_link.mov | ./ffmpeg -r 15 -i - -codec copy -hls_list_size 65535 -hls_time 2 "./video/live.m3u8"

  But I also need to draw some text on the video before recording. So I re-write the above command like

  openRTSP -D 10 -v -t -c -b 800000 rtsp://video_link.mov | ./ffmpeg -r 15 -i - -vf drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeMonoBold.ttf: text='Stack Overflow': fontcolor=white: fontsize=24: box=1: boxcolor=black: x=(w-text_w)/2: y=(h-text_h-line_h)/2" -codec copy -hls_list_size 65535 -hls_time 2 "./video/live.m3u8"

  But giving the error like,

  input #0, h264, from 'pipe:':
  
    Duration: N/A, bitrate: N/A
  
      Stream #0:0: Video: h264 (High), yuv420p, 704x480, 25 fps, 25 tbr, 1200k tbn, 50 tbc
  
  Filtergraph 'drawtext=fontfile=/usr/share/fonts/truetype/freefont/FreeMonoBold.ttf: text='Stack Overflow': fontcolor=white: fontsize=24: box=1: boxcolor=black: x=(w-text_w)/2: y=(h-text_h-line_h)/2' was defined for video output stream 0:0 but codec copy was selected.
  
  Filtering and streamcopy cannot be used together.

  I understand that both recording and drawing cannot perform together.
  Any method exist similar, to solve the above problem ?.

• Enhance text image in x264encoding

  23 août 2016, par useprxf

  I’m making use of x264 for remote desktop streaming. The goal is to achieve both low bitrate and high video quality within the computation budget. The current parameter set I used almost achieve this goal, but it fails in handling images with many texts (e.g. browsing websites schene). The text in image is blurred and affects the user experience.

  I think it’s the quantization in x264 that causes this. The quantization after DCT transform will eliminate high frequency sinals which mainly correspond to texts in image.

  So, my question is how to do improve the text quality in x264 encoding ?

  My idea : when the bitrate stays at a low level for a period of time,

  1. set crf to be 0 (lossless) ;
  2. encode current frame as an IDR frame and then send it ;
  3. recover the crf.

  Also, a flag should be used to prevent resending when bitrate keeps low for a long time. I haven’t try this method since I don’t know how to mark a frame as an IDR frame manully and then encode it.

• Combining fading text with reencoding video

  6 août 2020, par jmricker

  I would like to renecode video then put fading text on the resulting video. Both of the parts work seperaetely but when I put them both together I get the error

  


  Filtergraph 'scale=1280:720:force_original_aspect_ratio=decrease,setsar=1:1,pad=1280:720:(ow-iw)/2:(oh-ih)/2' 
was specified through the -vf/-af/-filter option for output stream 0:0, which is fed from a complex filtergraph.
-vf/-af/-filter and -filter_complex cannot be used together for the same stream.



  


  ffmpeg -i inputtest.mp4 -vcodec libx264 -crf 27 -preset veryfast -vf vscale=1280:720:force_original_aspect_ratio=decrease,setsar=1:1,pad=1280:720:(ow-iw)/2:(oh-ih)/2" \
    -filter_complex \
        "color=black@0:100x100,format=yuva444p[c]; [c][0]scale2ref[ct][mv0]; [ct]setsar=1,split=3[t1][t2][t3]; \
        [t1]drawtext=fontfile= Helvetica 400.ttf:text=\'Test1\':fontsize=40:fontcolor=white,fade=t=in:st=1:d=1:alpha=1,fade=t=out:st=5:d=1:alpha=1[txta1]; \
        [t2]drawtext=fontfile= Helvetica 400.ttf:text=\'Test2\':fontsize=40:fontcolor=white,fade=t=in:st=1:d=1:alpha=1,fade=t=out:st=10:d=1:alpha=1[txta2]; \
        [t3]drawtext=fontfile= Helvetica 400.ttf:text=\'Test3\':fontsize=40:fontcolor=white,fade=t=in:st=1:d=1:alpha=1,fade=t=out:st=15:d=1:alpha=1[txta3]; \
        [mv0][txta1]overlay=x=\'100\':y=\'200\':shortest=1[mv1]; \
        [mv1][txta2]overlay=x=\'300\':y=\'200\':shortest=1[mv2]; \
        [mv2][txta3]overlay=x=\'500\':y=\'200\':shortest=1" \
    output.mp4



  


  Any idea of what I can do to make this happen ? If nothing else I can separate the two commands but I thought putting together would make the encoding more efficient.