Recherche avancée

Sur d’autres sites (11990)

• How I Can Run Some Part of the Code on another EC2 instance

  4 mai 2021, par Not A Bot — ffmpeg

  I have a NodeJS application running on an EC2 instance, which has some feature where users can record multiple videos.

  


  When the user logs out I am using ffmpeg(version 4.2.4) to combine all those videos into single a video.

  


  I am recording the video in WEBM format, and the final single video should be in MP4 format.

  


  Suppose the user has recorded 3 videos of 10 minutes each, then in last when the user logs out, all these 3 videos should be combined into the single video on length of 30 minutes.

  


  Now everything is working fine, but the CPU usage is high when all the conversation and concatenation are going around.

  


  CPU usage is sometimes as high as 60-70%

  


  The process I am following is

  


  


  1. Convert the webm file to the mp4 file.

     


     ffmpeg -i input_file.webm -c:v copy -c:a copy output_file.mp4


     


  


  2. Convert MP4 to ts(Transport Stream) file.

     


      ffmpeg -i output_file.mp4 -c copy -bsf:v h264_mp4toannexb -f mpegts OUTPUT_MP4.ts


     


     I am following this process to concatenate all the mp4 files into one.

     


  


  3. Concatenate files

     


     ffmpeg -i "concat:OUTPUT_MP4_1.ts|OUTPUT_MP4_2.ts|OUTPUT_MP4_3.ts" -c copy -bsf:a aac_adtstoasc FINAL_MP4_SINGLE_FILE.mp4


     


  


  


  All this process is time-consuming(but not a priority), however, this process is taking CPU usage a lot.

  


  The server can crash or become slow if there are many users on my application and may video conversion is going on.

  


  Now, my question is how I can run this conversion process on the dedicated EC2 instance where only conversions can happen and not any other work, from the same code that is running on the first EC2 instance.

  


• avcodec/proresdec : rename dsp part for 10b and check dspinit for supported bits...

  17 novembre 2018, par Martin Vignali

  avcodec/proresdec : rename dsp part for 10b and check dspinit for supported bits per raw sample
  based on patch by Kieran Kunhya
  

  • [DH] libavcodec/proresdec2.c
  • [DH] libavcodec/proresdsp.c
  • [DH] libavcodec/proresdsp.h
  • [DH] libavcodec/simple_idct.c
  • [DH] libavcodec/simple_idct.h
  • [DH] libavcodec/tests/dct.c
  • [DH] libavcodec/x86/proresdsp_init.c

• can not split video into smaller part

  13 juillet 2021, par david

  I want to split some videos into 2 sec parts and for this aim, I am trying to use the following code :

  


  ffmpeg -i invid.264 -threads 3 -vcodec copy -f segment -segment_time 2 cam_out_h264%04d.264


  


  but it cannot work properly and it produces this error in the command line :

  


  [h264 @ 000001b4caf5f580] Stream #0: not enough frames to estimate rate; consider increasing probesize
Input #0, h264, from 'output.264':
  Duration: N/A, bitrate: N/A
  Stream #0:0: Video: h264 (High), yuv420p(progressive), 1920x1080 [SAR 1:1 DAR 16:9], 30 fps, 30 tbr, 1200k tbn, 60 tbc
[segment @ 000001b4cb0e3600] Opening 'cam_out_h2640000.h264' for writing
Output #0, segment, to 'cam_out_h264%04d.h264':
  Metadata:
    encoder         : Lavf58.77.100
  Stream #0:0: Video: h264 (High), yuv420p(progressive), 1920x1080 [SAR 1:1 DAR 16:9], q=2-31, 30 fps, 30 tbr, 30 tbn, 30 tbc
Stream mapping:
  Stream #0:0 -> #0:0 (copy)
Press [q] to stop, [?] for help
[segment @ 000001b4cb0e3600] Timestamps are unset in a packet for stream 0. This is deprecated and will stop working in the future. Fix your code to set the timestamps properly
frame=  600 fps=0.0 q=-1.0 Lsize=N/A time=00:00:19.93 bitrate=N/A speed= 234x
video:19039kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown


  


  what is the problem ? this video has 600 frames and it is 19.93 sec. hence, it should not have any problem for splitting into 2 sec videos.do you know what is the problem ? please help me with this issue.