Recherche avancée

Sur d’autres sites (7420)

• FFmpeg continues to process after time specified at "-to"

  4 août 2020, par Yamahabest

  I have a video, where I want to cut a part from the beginning, and from the end. And I want to apply some fade ins/fade outs, and add some text.

  


  So, I came up with the following syntax :

  


  -ss 10 -to 40 
-i "D:\DATA\Software\VideoProcessor_Files\20171015 Zelhem Tandem Frans met Mirthe.MP4" 
-loop 1 -i "Input_Files\logo maurik large.png" 
-loop 1 -i "Input_Files\logo maurik small.png" 
-filter_complex "
    color=0x7F7F7F@0.95:1920x1080[grey_for_fade_out];
    [grey_for_fade_out]fade=t=out:st=12:d=2:alpha=1[grey_fade_out];
    [0:v][grey_fade_out]overlay[video_grey_fade_out];
    color=0x7F7F7F@0.95:1920x1080[grey_for_fade_in];
    [grey_for_fade_in]fade=t=in:st=37:d=2:alpha=1[grey_fade_in];
    [video_grey_fade_out][grey_fade_in]overlay[video_grey_fade_out_in];
    [1:v]fade=t=out:st=13:d=2:alpha=1[over];
    [over]scale=iw/1.5:-1[scaled];
    [video_grey_fade_out_in][scaled]overlay=(main_w-overlay_w)/2:(main_h-overlay_h)/10[video_grey_fade_out_in_logo];
    [1:v]fade=t=in:st=36:d=2:alpha=1[over2];
    [over2]scale=iw/1.5:-1[scaled2];
    [video_grey_fade_out_in_logo][scaled2]overlay=(main_w-overlay_w)/2:(main_h-overlay_h)/2[video_grey_fade_out_in_logo2];
    [2:v]colorchannelmixer=aa=0.5,fade=t=in:st=14:d=2:alpha=1,fade=t=out:st=35:d=2:alpha=1[over3];
    [over3]scale=iw/10:-1[scaled3];
    [video_grey_fade_out_in_logo2][scaled3]overlay=10:10[video_complete];
    [video_complete]drawtext=fontfile=Input_Files/Sansation-Bold.ttf:text='Tandemvlucht met Mirthe':fontsize=96:fontcolor=white:alpha='if(lt(t,11),1,(2-(t-11))/2)':x=(w-text_w)/2:y=((h-text_h)/2)+125,drawtext=fontfile=Input_Files/Sansation-Bold.ttf:text='Zeddam':fontsize=96:fontcolor=white:alpha='if(lt(t,11),1,(2-(t-11))/2)':x=(w-text_w)/2:y=((h-text_h)/2)+250,drawtext=fontfile=Input_Files/Sansation-Bold.ttf:text='4 augustus 2020':fontsize=96:fontcolor=white:alpha='if(lt(t,11),1,(2-(t-11))/2)':x=(w-text_w)/2:y=((h-text_h)/2)+375,drawtext=fontfile=Input_Files/Sansation-Bold.ttf:text='Ook een keer meevliegen?':fontsize=96:fontcolor=white:alpha='if(lt(t,37),0,(t-37)/2)':x=(w-text_w)/2:y=((h-text_h)/6),drawtext=fontfile=Input_Files/Sansation-Bold.ttf:text='Of bel 085 - 049 55 69':fontsize=96:fontcolor=white:alpha='if(lt(t,37),0,(t-37)/2)':x=(w-text_w)/2:y=((h-text_h)/2)+350"
-preset medium 
-crf 18 
-c:a copy 
-y ".\Output_Files\Video\Zeddam\2020-08-04\Mirthe\27ed390a-8497-4550-b93f-4f87d9f2c9f0\MP_Tandemvlucht met_Mirthe_Zeddam_2020-08-04.mp4"


  


  I am quite sure this has worked in the past, but now FFmpeg just keeps on processing endlessly. If I then stop the FFmpeg process, and look at the resulting file, I see that the last frame (of the end of the specified period) just keeps on duplicating.

  


  The drop counter in the console output of FFmpeg also starts increasing at the end of the specified period :

  


  frame=  987 fps= 15 q=-1.0 Lsize=   31357kB time=00:00:41.04 bitrate=6259.0kbits/s dup=0 drop=10 speed=0.644x


  


  I am kind of lost on why this doesn't work anymore. I might have upgraded the FFmpeg executable in the mean time. Maybe my syntax was/is not correct, but I believe it just worked.

  


  It has to be in the complex filter, because when I remove that, it works alright.
It is not in the drawtext part of the complex filter, because it still occurs when I remove that. And when I only do the drawtext, FFmpeg stops correctly at the specified time.

  


  I have tried this, but then it still occurs :

  


  -filter_complex "
    color=0x7F7F7F@0.95:1920x1080[grey_for_fade_out];
    [grey_for_fade_out]fade=t=out:st=12:d=2:alpha=1[grey_fade_out];
    [0:v][grey_fade_out]overlay[video_grey_fade_out];
    color=0x7F7F7F@0.95:1920x1080[grey_for_fade_in];
    [grey_for_fade_in]fade=t=in:st=37:d=2:alpha=1[grey_fade_in];
    [video_grey_fade_out][grey_fade_in]overlay"


  


  Also with this, it still occurs :

  


  -filter_complex "
    [1:v]fade=t=out:st=13:d=2:alpha=1[over];
    [over]scale=iw/1.5:-1[scaled];
    [0:v][scaled]overlay=(main_w-overlay_w)/2:(main_h-overlay_h)/10[video_grey_fade_out_in_logo];
    [1:v]fade=t=in:st=36:d=2:alpha=1[over2];
    [over2]scale=iw/1.5:-1[scaled2];
    [video_grey_fade_out_in_logo][scaled2]overlay=(main_w-overlay_w)/2:(main_h-overlay_h)/2[video_grey_fade_out_in_logo2];
    [2:v]colorchannelmixer=aa=0.5,fade=t=in:st=14:d=2:alpha=1,fade=t=out:st=35:d=2:alpha=1[over3];
    [over3]scale=iw/10:-1[scaled3];
    [video_grey_fade_out_in_logo2][scaled3]overlay=10:10"


  


  I just don't understand. All my fade-ins/-outs seem to be within the specified range :

  


  

  • fade=t=out:st=12:d=2 : start at 12 seconds, duration 2 seconds. This ends at 14 seconds, which is smaller than 40 seconds.

  


  • fade=t=in:st=37:d=2 : start at 37 seconds, duration 2 seconds. This ends at 39 seconds, which is smaller than 40 seconds.

  


  • fade=t=out:st=13:d=2 : start at 13 seconds, duration 2 seconds. This ends at 15 seconds, which is smaller than 40 seconds.

  


  • fade=t=in:st=36:d=2 : start at 36 seconds, duration 2 seconds. This ends at 38 seconds, which is smaller than 40 seconds.

  


  


  It is just like some sequence is not ended properly, which is causing FFmpeg to just continue.

  


• ffmpeg -ss then apply filter then concat producing timestamp errors

  6 août 2020, par Bob Ramsey

  Using ffmpeg, I have split a file into multiple parts using -ss. Then I apply a filter to some of the files, then concat the files back together. When I do that, I get : Non-monotonous DTS in output stream 0:0 ; previous : 341334, current : 340526 ; changing to 341335. This may result in incorrect timestamps in the output file. The output file plays, but there are noticeable skips where the files are joined.

  


  Here's how I am splitting the file :

  


  ffmpeg -i full_source.mp4 -ss 0 -to 14.264250 -c copy  01-plain.mp4
ffmpeg -i full_source.mp4 -ss 14.264250 -to 18.435083 -c copy  01-filtered.mp4

ffmpeg -i full_source.mp4 -ss 18.435083 -to 29.988292 -c copy  02-plain.mp4
ffmpeg -i full_source.mp4 -ss 29.988292 -to 31.865167 -c copy  02-filtered.mp4
...
ffmpeg -i full_source.mp4 -ss 0 -to 14.264250 -c copy  10-plain.mp4
ffmpeg -i full_source.mp4 -ss 234.484203 -to 300.000 -c copy  10-filtered.mp4


  


  Then I apply a different drawtext filter on each of the 10 filtered files and save them with a new name, like :

  


  ffmpeg -hide_banner -loglevel warning -y -i 01-filtered.mp4 -filter_complex "drawtext=fontfile=calibri.ttf:fontsize=24:fontcolor=white:x=300:y=500:text='hello world'" -crf 15 01-filtered-complete.mp4


  


  Finally, I join all of the plain and complete files back together like this :

  


  ffmpeg  -f concat -safe 0 -i mylist.txt -c:a copy -c:v copy  outfile.mp4


  


  And that's where the timing error comes in. I've tried adding -vsync drop in the concat command, but that didn't really work either. Same version of ffmpeg does the split, the filter, and the concat. I've tried different versions, everything from 20170519 to one from May 2020 with the same result. Always making sure that all three steps are done by the same version of ffmpeg.

  


  The only thing I can see is that ffprobe shows a duration of 14.27 for 01-plain.mp4 when it should be 14.264250. All of the other files show a similar rounding difference. The files are 23.98 fps. If I do all of my filters in really long command without splitting the file, I can use the more precise numbers with no problem. It just takes 10 times as long. This is all scripted, it happens a couple of hundred times a day and time is money, so I can't take 10 times as long to do each file.

  


  Any ideas ? Thanks in advance !

  


• FFmpeg start slowly with multiple instance on nvidia T4 GPU

  3 août 2020, par wyWang

  


  environment：

  


  

  1. OS:Centos7

  


  2. FFMPEG version : 4.2

  


  3. GPU : NVIDIA Tesla T4

  


  


  


  I have a requirement to work with multiple FFMPEG instance,command like this :

  


  ffmpeg -y -loglevel info -hwaccel cuvid -hwaccel_output_format cuda -c:v h264_cuvid \
    -i http://xx/xxxx.mp4 \
    -c:v  h264_nvenc  -ss 0 -t 600 out.mp4


  


  and the output like this :

  


  ffmpeg version 4.2.git Copyright (c) 2000-2020 the FFmpeg developers
  built with gcc 4.8.5 (GCC) 20150623 (Red Hat 4.8.5-39)
  configuration: --enable-cuda-nvcc --enable-cuda --enable-cuvid --enable-nvenc --enable-libnpp --prefix=/snbeauty//platform/centos --pkg-config-flags=--static --extra-cflags='-I/snbeauty//platform/centos/include -I/usr/local/cuda/include/ -Werror=implicit-function-declaration -std=gnu99' --extra-ldflags='-L/snbeauty//platform/centos/lib64 -L/snbeauty//platform/centos/lib -L/usr/local/lib64 -L/usr/lib64 -L/usr/local/cuda/lib64 -L/lib64' --extra-libs='-lpthread -lm -lEGL -lGL -lpng -lYNEffect -ltracker -lcuda -lcudart' --bindir=/snbeauty//platform/centos/bin --enable-static --enable-nonfree --enable-libfreetype --enable-libfribidi --enable-libfontconfig --enable-filters --enable-version3 --enable-gpl --enable-libass --enable-libfdk_aac --enable-encoder=libfdk_aac --enable-opengl --enable-filter=xiu --enable-libmp3lame --enable-libopus --enable-libvorbis --enable-libvpx --enable-libx264 --enable-libx265 --enable-libtheora --enable-libwebp --enable-frei0r --enable-openssl --enable-gray --enable-libxml2 --enable-libxvid
  libavutil      56. 42.100 / 56. 42.100
  libavcodec     58. 73.102 / 58. 73.102
  libavformat    58. 39.101 / 58. 39.101
  libavdevice    58.  9.103 / 58.  9.103
  libavfilter     7. 77.100 /  7. 77.100
  libswscale      5.  6.100 /  5.  6.100
  libswresample   3.  6.100 /  3.  6.100
  libpostproc    55.  6.100 / 55.  6.100
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'http://xxx/xxxx.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf58.29.100
  Duration: 00:38:52.05, start: 0.000000, bitrate: 5370 kb/s
    Stream #0:0(und): Video: h264 (Main) (avc1 / 0x31637661), yuv420p, 1920x1080, 5245 kb/s, 25 fps, 25 tbr, 90k tbn, 50 tbc (default)
    Metadata:
      handler_name    : VideoHandler
    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 119 kb/s (default)
    Metadata:
      handler_name    : SoundHandler
Stream mapping:
  Stream #0:0 -> #0:0 (h264 (h264_cuvid) -> h264 (h264_nvenc))
  Stream #0:1 -> #0:1 (aac (native) -> aac (native))
Press [q] to stop, [?] for help
Output #0, mp4, to 'out.mp4':
  Metadata:
    major_brand     : isom
    minor_version   : 512
    compatible_brands: isomiso2avc1mp41
    encoder         : Lavf58.39.101
    Stream #0:0(und): Video: h264 (h264_nvenc) (Main) (avc1 / 0x31637661), cuda, 1920x1080 [SAR 1:1 DAR 16:9], q=-1--1, 2000 kb/s, 25 fps, 12800 tbn, 25 tbc (default)
    Metadata:
      handler_name    : VideoHandler
      encoder         : Lavc58.73.102 h264_nvenc
    Side data:
      cpb: bitrate max/min/avg: 0/0/2000000 buffer size: 4000000 vbv_delay: N/A
    Stream #0:1(und): Audio: aac (LC) (mp4a / 0x6134706D), 44100 Hz, stereo, fltp, 128 kb/s (default)
    Metadata:
      handler_name    : SoundHandler
      encoder         : Lavc58.73.102 aac
frame=15000 fps=336 q=26.0 Lsize=  152462kB time=00:10:00.00 bitrate=2081.6kbits/s speed=13.4x    
video:142668kB audio:9418kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.247139%
[aac @ 0x32437c0] Qavg: 168.657

real    0m45.568s
user    0m29.034s
sys 0m2.961s


  


  is very fast

  


  But when I try concurrent 10 processes to run this command with different input url,ffmpeg start is very slowly. Then I found FFmpeg is slowly in this step :

  


  Stream mapping:
  Stream #0:0 -> #0:0 (h264 (h264_cuvid) -> h264 (h264_nvenc))
  Stream #0:1 -> #0:1 (aac (native) -> aac (native))


  


  When the video is very short(like 0.5s or 1s), FFmpeg takes a long time to start.The actual codec takes less time than the boot time.

  


  I would like to ask what causes this problem ? (Nvidia GPU limit or FFmpeg) .And any way to solve it.

  


  Thanks !