Recherche avancée

Sur d’autres sites (12873)

• How do I alter my FFMPEG command to make my HTTP Live Streams more efficient ?

  17 octobre 2014, par Robert

  I want to reduce the muxing overhead when creating .ts files using FFMPEG.

  Im using FFMPEG to create a series of transport stream files used for HTTP live streaming.

  ./ffmpeg -i myInputFile.ismv \
  
           -vcodec copy \
  
           -acodec copy \
  
           -bsf h264_mp4toannexb \
  
           -map 0 \
  
           -f segment \
  
           -segment_time 10\
  
           -segment_list_size 999999 \
  
           -segment_list output/myVarientPlaylist.m3u8 \
  
           -segment_format mpegts \
  
           output/myAudioVideoFile-%04d.ts

  My input is in ismv format and contains a video and audio stream :

  Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 320x240, 348 kb/s, 29.97 tbr, 10000k tbn, 59.94 tbc
  
  Stream #0:1(und): Audio: aac (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 63 kb/s

  There is an issues related to muxing that is causing a large amout of overhead to be added to the streams. This is how the issue was described to me for the audio :

  

  So for a given aac stream, the overhead will be 88% (since 200 bytes will map to 2 x 188 byte packets).

  For video, the iframe packets are quite large, so they translate nicely into .ts packets, however, the diffs can be as small as an audio packet, therefore they suffer from the same issue.

  The solution is to combine several aac packets into one larger stream before packaging them into .ts. Is this possible out of the box with FFMPEG ?

• can't read mp4 in opencv3.2 (ubuntu, python3)

  28 mars 2017, par lhk

  I’ve ran into a problem with my opencv installation, it is unable to open an mp4 video. My system is ubuntu 16.04, 64bit, opencv3.2 used from python 3.5.

  VideoCapture.read returns False and None.

  There are other questions with this problem, but they target different platforms or different opencv versions.

  Apparently, I’m missing the proper codec.
  So I ran make uninstall from my build directory, purged opencv* with apt and built from source again. This time making sure that ffmpeg was installed before the compilation.

  Here are my steps :

  • clone opencv and opencv_contrib
  • cd opencv/
  • mkdir build
  • cd build
  • cmake -D CMAKE_BUILD_TYPE=RELEASE     -D CMAKE_INSTALL_PREFIX=/usr/local     -D INSTALL_PYTHON_EXAMPLES=ON     -D INSTALL_C_EXAMPLES=OFF     -D OPENCV_EXTRA_MODULES_PATH=../../opencv_contrib/modules     -D BUILD_EXAMPLES=ON ..
  • make -j 8
  • sudo make install

  I checked the output of cmake, ffmpeg is there :

  Video I/O:
  
  --     DC1394 1.x:                  NO
  
  --     DC1394 2.x:                  NO
  
  --     FFMPEG:                      YES
  
  --       avcodec:                   YES (ver 56.60.100)
  
  --       avformat:                  YES (ver 56.40.101)
  
  --       avutil:                    YES (ver 54.31.100)
  
  --       swscale:                   YES (ver 3.1.101)
  
  --       avresample:                NO
  
  --     GStreamer:                   NO
  
  --     OpenNI:                      NO
  
  --     OpenNI PrimeSensor Modules:  NO
  
  --     OpenNI2:                     NO
  
  --     PvAPI:                       NO
  
  --     GigEVisionSDK:               NO
  
  --     Aravis SDK:                  NO
  
  --     UniCap:                      NO
  
  --     UniCap ucil:                 NO
  
  --     V4L/V4L2:                    NO/YES
  
  --     XIMEA:                       NO
  
  --     Xine:                        NO
  
  --     gPhoto2:                     NO

  But the problem persists. How can I fix this ?

  UPDATE

  I had to manually remove some .so files from /usr/local.

  Then I installed all avi related codecs I could find.
  https://wiki.ubuntuusers.de/Codecs/
  plus libavcodec-extra and ffmpeg

  Then I recompiled and now it works.

• ffmpeg for archival and convertibility

  27 juillet 2021, par Satya

  I've got a couple hundred gigs of *.dv files. I'd like to convert them to H.264 or something else or even leave them alone. The purpose is archival, with an eye to maximum convertibility especially to DVD. The content is family videos.

  


  Would this be fine ?

  


  ffmpeg -i input.dv \
    -c:v libx264 -preset slower \
    -crf 17 \
    -pix_fmt yuv420p \
    output.mp4


  


  I went with the slower preset because encoding time isn't an issue and I'd like a smaller file size. crf 17 is for least-lossy while being widely playable. I read somewhere that yuv420p is needed for some Quicktime players.

  


  Should I throw in -c:a aac for AAC audio ? The audio is voice only, no need for music-hall quality.

  


  I looked at https://trac.ffmpeg.org/wiki/Encode/H.264 for previous research and that's where I got those settings, but it is silent on the audio settings.

  


  Edited : My priorities, in order of importance, are :

  


  

  1. Compatibility

  


  2. Losslessness (doesn't have to be 100% lossless, hence crf of 17 and not 0)

  


  3. File size

  


  


  Most of the input files say this :

  


  [lavf] stream 0: video (dvvideo), -vid 0
[lavf] stream 1: audio (pcm_s16le), -aid 0
VIDEO:  [dvsd]  720x480  0bpp  29.970 fps  25000.0 kbps (3051.8 kbyte/s)
Selected video codec: [ffdv] vfm: ffmpeg (FFmpeg DV)
AUDIO: 32000 Hz, 2 ch, s16le, 1024.0 kbit/100.00% (ratio: 128000->128000)
Selected audio codec: [pcm] afm: pcm (Uncompressed PCM)


  


  Output from ffmpeg :

  


  Stream #0:1: Audio: pcm_s16le, 32000 Hz, stereo, s16, 1024 kb/s