Recherche avancée

Sur d’autres sites (5335)

• How can I avoid an ffmpeg out of memory error

  21 mars 2021, par Silentfury

  I use ffmpeg with complex filtering. Input are different sets of FULLHD surveilance camera videos each 10 to 15 seconds long. Set size (number of videos per set) varies. To remove unchanged frames I apply mpdecimate. To avoid being triggered by moving bushes but still keep objects I want to remain, I apply a complex filter :

  


  

  • split the video (the original and a dummy to detect motion/stills)

  


  • scale the dummy down (so the 8x8-block-metric of mpdecimate matches the size of moving objects I want to keep)

  


  • add white boxes to dummy to mask unintendedly moving objects

  


  • apply mpdecimate to dummy to remove non-changing frames

  


  • scale dummy back to original size

  


  • overlay the remaining frames of dummy with matching frames of original

  


  


  All this works fine if the number of input videos is small (less than 100). The memory consupmtion of the ffmpeg process varies somewhere between 2GiB and 5GiB.

  


  If the number of input files gets larger (say 200), the memory consumption suddenly jumps to insane numbers until memory (32GiB plus 33GiB swap) runs out and ffmpeg gets killed. I can not predict if and why this happens. I have one example, where a set of 340 videos worked using 6GiB. Any other set above 100 videos I tried eats all RAM in under two minutes and dies.

  


  There is no particular error message from ffmpeg.

  


  dmesg says :

  


  Out of memory: Kill process 29173 (ffmpeg)
Killed process 29173 (ffmpeg) total-vm:66707800kB


  


  My ffmpeg command :

  


  ffmpeg -f concat -safe 0 -i vidlist -vf 'split=2[full][masked];[masked]scale=w=iw/4:h=ih/4,drawbox=w=51:h=153:x=101:y=0:t=fill:c=white,drawbox=w=74:h=67:x=86:y=49:t=fill:c=white,drawbox=w=51:h=149:x=258:y=0:t=fill:c=white,drawbox=w=13:h=20:x=214:y=103:t=fill:c=white,drawbox=w=29:h=54:x=429:y=40:t=fill:c=white,drawbox=w=35:h=49:x=360:y=111:t=fill:c=white,drawbox=w=26:h=54:x=304:y=92:t=fill:c=white,drawbox=w=48:h=27:x=356:y=105:t=fill:c=white,drawbox=w=30:h=27:x=188:y=124:t=fill:c=white,drawbox=w=50:h=54:x=371:y=7:t=fill:c=white,drawbox=w=18:h=38:x=248:y=107:t=fill:c=white,drawbox=w=21:h=51:x=242:y=33:t=fill:c=white,mpdecimate=hi=64*80:lo=64*40:frac=0.001,scale=w=iw*4:h=ih*4[deduped];[deduped][full]overlay=shortest=1,setpts=N/(15*TB),mpdecimate=hi=64*80:lo=64*50:frac=0.001,setpts=N/(15*TB)' -r 15 -c:v libx265 -preset slower -crf 37 -pix_fmt yuv420p -an result.mkv


  


  ffmpeg version 4.1.6

  


  Debian 4.19.171-2

  


  I hope that my filter can be tuned in some way that achieves the same result but doesn't eat RAM that much - but I have no clue how. Within reasonable limits, I wouldn't mind if processing time suffers. Any hints appreciated.

  


• Converting MTS to MP4 using FFmpeg results in no/missing audio

  17 mars 2021, par tenninebt

  I have a few videos coming from a Sony Nex-5N. Basically they are "compiled" in an AVCHD. The format that is inside is MTS. I'm able to convert the videos losslessly but the resulting MP4 has no audio.

  


  ffmpeg -i 20201205150626.MTS -c:v copy -c:a copy -an -f mp4 20201205150626.MP4 


  


  The result is no audio. Then I tried to select explicitly the audio track using -map 0:a:1 but it says that it does not matches no streams. Please have a look at the following extract :

  


  > ffmpeg -i 20201205150626.MTS -c:v copy -c:a copy -map 0:a:1 -an -f mp4 20201205150626.MP4 
ffmpeg version 4.3.2 Copyright (c) 2000-2021 the FFmpeg developers
  built with Apple clang version 12.0.0 (clang-1200.0.32.29)
  configuration: --prefix=/usr/local/Cellar/ffmpeg/4.3.2 --enable-shared --enable-pthreads --enable-version3 --enable-avresample --cc=clang --host-cflags= --host-ldflags= --enable-ffplay --enable-gnutls --enable-gpl --enable-libaom --enable-libbluray --enable-libdav1d --enable-libmp3lame --enable-libopus --enable-librav1e --enable-librubberband --enable-libsnappy --enable-libsrt --enable-libtesseract --enable-libtheora --enable-libvidstab --enable-libvorbis --enable-libvpx --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libxvid --enable-lzma --enable-libfontconfig --enable-libfreetype --enable-frei0r --enable-libass --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-librtmp --enable-libspeex --enable-libsoxr --enable-libzmq --enable-libzimg --disable-libjack --disable-indev=jack --enable-videotoolbox
  libavutil      56. 51.100 / 56. 51.100
  libavcodec     58. 91.100 / 58. 91.100
  libavformat    58. 45.100 / 58. 45.100
  libavdevice    58. 10.100 / 58. 10.100
  libavfilter     7. 85.100 /  7. 85.100
  libavresample   4.  0.  0 /  4.  0.  0
  libswscale      5.  7.100 /  5.  7.100
  libswresample   3.  7.100 /  3.  7.100
  libpostproc    55.  7.100 / 55.  7.100
Input #0, mpegts, from '20201205150626.MTS':
  Duration: 00:00:09.60, start: 1.020000, bitrate: 26296 kb/s
  Program 1 
    Stream #0:0[0x1011]: Video: h264 (High) (HDPR / 0x52504448), yuv420p(progressive), 1920x1080 [SAR 1:1 DAR 16:9], 50 fps, 50 tbr, 90k tbn, 100 tbc
    Stream #0:1[0x1100]: Audio: ac3 (AC-3 / 0x332D4341), 48000 Hz, stereo, fltp, 256 kb/s
    Stream #0:2[0x1200]: Subtitle: hdmv_pgs_subtitle ([144][0][0][0] / 0x0090), 1920x1080
Stream map '0:a:1' matches no streams.
To ignore this, add a trailing '?' to the map.



  


  I can see an audio stream, and its index is 1. I have no clue what is happening.

  


• Using an actual audio recording to filter out noise from a video

  9 mars 2021, par user2751530

  I use my laptop (Ubuntu 18.04 LTS derivative on a Dell XPS13) for recording videos (these are just narrated presentations) using OBS. After a presentation is done (.flv format), I process it using ffmpeg using filters that try to reduce background noise, reduce the size of the video, change encoding to .mp4, insert a watermark, etc. Over several months, this system has worked well.

  


  However, my laptop is now beginning to show its age (it is 4 years old). That means that the fan becomes loud - loud enough to notice in a recording, not loud enough to notice when you are working. So, even after filtering for low frequency in ffmpeg, there are clicking and other type of sounds that are left in the video. I am a scientist, though not an audio/video expert. So, I was thinking - is it possible for me to simply record the noise coming out of my machine when I am not presenting, and then use that recording to filter out the noise that my machine makes during the presentation ?

  


  Blanket approaches like filtering out certain ranges of the audio spectrum, etc. are unlikely to work, as the power spectrum of the noise likely has many peaks, and these are likely to extend into human voice range as well (I can hear them). Further, this is a moving target - the laptop is aging and in any case, the amount and type of noise it makes depends on the load and how long it has been on. Algorithm :

  


  

  1. Record actual computer noise (with the added bonus of background noise) while I am not recording. Ideally, just before starting to record the presentation. This could take the form of a 1-2 minute audio sample.

  


  2. Record the presentation on OBS.

  


  3. Use 1 as a filter to get rid of noise in 2. I imagine it would involve doing a Fourier analysis of 1, and then removing those peaks from the spectrum of 2 at each time epoch.

  


  


  I have looked into sox, which is what people somewhat flippantly point you to without giving any details. I do not know how to separate out audio channels from a video and then interleave them back together (not an expert on the software here). Other than RTFM, is there any helpful advice anyone could offer ? I have searched, but have not been able to find a HOWTO. I expect that that is probably the fault of my search since I refuse to believe that this is a new idea - it is a standard method used in many fields to get rid of noise, including astronomy.