Recherche avancée

Sur d’autres sites (8944)

• Accord.Video.FFMpeg : Problem of duration difference between recording time and video time

  5 mars 2020, par ckrby

  I want to save the footage I get from the webcam as a video in webM (Vp8) format. I use the Accord library. But when recording, there is almost 2-3 times the difference between the time spent and the video time. The code I use and some debug outputs are below.

  Accord.Video.FFMpeg 3.8.2-Alpha (double pts/dts bug fixed version)
  Video Save Format : VP8 (.WebM)

  Record Start Settings :

  videoCodec = VideoCodec.Vp8;    
  
  bitrate : 1250000   
  
  framerate :10   
  
  height :720 
  
  width: 1280 
  
  videoSource.VideoResolution: 1280x720, 30 fps (30 max fps), 24 bpp  
  
  FileWriter.BitRate = bitRate;   
  
  FileWriter.FrameRate = new Accord.Math.Rational((int)framerate );   
  
  FileWriter.Width = width;   
  
  FileWriter.Height = height; 
  
  FileWriter.VideoCodec = videoCodec; 
  
  FileWriter.Open(FileName);  
  
  _recording = true;  
  
  StartTimeStamp = DateTime.Now;  

  newFrame Event

  private void VideoSourcePlayer1_NewFrame(object sender, ref Bitmap image)
  
  {
  
      if (_recording)
  
      {
  
          try
  
          {
  
              Bitmap CurrentBitmap = image;
  
              DateTime CurrentTimeStamp = DateTime.Now;
  
              TimeSpan CurrentTimeSpan = StartTimeStamp - CurrentTimeStamp;
  
              Console.WriteLine("Video Write Timespan: " + CurrentTimeSpan.ToString());
  
              FileWriter.WriteVideoFrame(CurrentBitmap, CurrentTimeSpan);
  
              // FileWriter.WriteVideoFrame(image);  
  
          }
  
          catch (Exception ex)
  
          {
  
              Console.WriteLine("VideoSourcePlayer1_NewFrame catch: " + ex.Message);
  
          }
  
      }
  
  }

  Note : It is the same issue with timespan or without timespan writing.

  In the 122 second (02.02 min) period, 50 seconds of video recording is created. I couldn’t solve it.

  Little Part of the debug output :

  Video Write Timespan : -00:01:55.4107365
  pts:4,730000e+004 pts_time:47,3 dts:4,730000e+004 dts_time:47,3 duration:1,000000e+002 duration_time:0,1
  Video Write Timespan : -00:01:55.6136263
  pts:4,740000e+004 pts_time:47,4 dts:4,740000e+004 dts_time:47,4 duration:1,000000e+002 duration_time:0,1
  Video Write Timespan : -00:01:55.8354960
  pts:4,750000e+004 pts_time:47,5 dts:4,750000e+004 dts_time:47,5 duration:1,000000e+002 duration_time:0,1
  Video Write Timespan : -00:01:56.5161088
  pts:4,760000e+004 pts_time:47,6 dts:4,760000e+004 dts_time:47,6 duration:1,000000e+002 duration_time:0,1
  Video Write Timespan : -00:01:56.7309841
  pts:4,770000e+004 pts_time:47,7 dts:4,770000e+004 dts_time:47,7 duration:1,000000e+002 duration_time:0,1
  Video Write Timespan : -00:01:56.9758413
  pts:4,780000e+004 pts_time:47,8 dts:4,780000e+004 dts_time:47,8 duration:1,000000e+002 duration_time:0,1
  Video Write Timespan : -00:01:57.1947200
  pts:4,790000e+004 pts_time:47,9 dts:4,790000e+004 dts_time:47,9 duration:1,000000e+002 duration_time:0,1
  Video Write Timespan : -00:01:57.4375834
  pts:4,800000e+004 pts_time:48 dts:4,800000e+004 dts_time:48 duration:1,000000e+002 duration_time:0,1
  Video Write Timespan : -00:01:57.5804983

  The last two outputs :

  Video Write Timespan : -00:02:02.7505463
  pts:5,000000e+004 pts_time : 50 dts:5,000000e+004 dts_time : 50 duration:1,000000e+002 duration_time:0,1
  Video Write Timespan : -00:02:02.9884038
  pts:5,010000e+004 pts_time : 50,1 dts:5,010000e+004 dts_time : 50,1 duration:1,000000e+002 duration_time:0,1

  Why is there a difference between timespan and dts/pts time ? Could the problem be due to this ?

  Note : When I set the framerate to 30 ; Another process 95 seconds (1.35 min) recording time is 16 seconds in the video.

  As the frame rate increases, the length of the video gets shorter and
  shorter than the actual value it should be.

  Output :

  Video Write Timespan : -00:01:35.1086178
  pts:1,606700e+004 pts_time : 16,067 dts:1,606700e+004 dts_time : 16,067 duration:3,300000e+001 duration_time:0,033
  Video Write Timespan : -00:01:35.2565302
  pts:1,610000e+004 pts_time : 16,1 dts:1,610000e+004 dts_time : 16,1 duration:3,300000e+001 duration_time:0,033
  Video Write Timespan : -00:01:35.3984494
  pts:1,613300e+004 pts_time : 16,133 dts:1,613300e+004 dts_time : 16,133 duration:3,300000e+001 duration_time:0,033

  So the video duration is like dts_time. Why is the video duration not
  value of timespan time, I still can’t find it ??

  Soory for my bad english & expression,

• muxing DVB subtitles into transport stream with ffmpeg

  28 novembre 2019, par user3439130

  So i have H264 video + AAC audio muxed inside transport stream (*.ts). To achive evenly spread PCR values i added muxrate tag to my command.

  ffmpeg -analyzeduration 20000000 -probesize 20M -loglevel verbose -i 1135084.m4v -i 1135084.m4a -c copy -muxrate 7982K -map 0:v:0 -map 1:a:0 -metadata:s:a:0 language=eng -metadata:s:a:0 tms_track_id=169451954 TEMP0.ts

  Since our client need DVB substitles we convert SRT to DVD with subtitle edit and then encode them to DVB with ffmpeg with this command :

  ffmpeg -analyzeduration 20000000 -probesize 20M -loglevel verbose -i TEMP0.ts -i out.da.idx -map 0 -c copy -muxrate 7992K -map 1:s -c:s:0 dvbsub -metadata:s:s:0 language=dan TEMP1.ts

  here is output from ffmpeg command :

  ffmpeg version git-2019-11-18-d831edc Copyright (c) 2000-2019 the FFmpeg developers
  
    built with gcc 9.2.1 (GCC) 20191010
  
    configuration: --enable-gpl --enable-version3 --enable-sdl2 --enable-fontconfig --enable-gnutls --enable-iconv --enable-libass --enable-libdav1d --enable-libbluray --enable-libfreetype --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libtheora --enable-libtwolame --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libzimg --enable-lzma --enable-zlib --enable-gmp --enable-libvidstab --enable-libvorbis --enable-libvo-amrwbenc --enable-libmysofa --enable-libspeex --enable-libxvid --enable-libaom --enable-libmfx --enable-ffnvcodec --enable-cuvid --enable-d3d11va --enable-nvenc --enable-nvdec --enable-dxva2 --enable-avisynth --enable-libopenmpt --enable-amf
  
    libavutil      56. 36.100 / 56. 36.100
  
    libavcodec     58. 62.100 / 58. 62.100
  
    libavformat    58. 35.100 / 58. 35.100
  
    libavdevice    58.  9.101 / 58.  9.101
  
    libavfilter     7. 66.100 /  7. 66.100
  
    libswscale      5.  6.100 /  5.  6.100
  
    libswresample   3.  6.100 /  3.  6.100
  
    libpostproc    55.  6.100 / 55.  6.100
  
  [h264 @ 000002bf7460d400] non-existing SPS 0 referenced in buffering period
  
  [h264 @ 000002bf7460d400] SPS unavailable in decode_picture_timing
  
  [h264 @ 000002bf7460d400] non-existing SPS 0 referenced in buffering period
  
  [h264 @ 000002bf7460d400] SPS unavailable in decode_picture_timing
  
  [h264 @ 000002bf7460d400] Reinit context to 1920x1088, pix_fmt: yuv420p
  
  [mpegts @ 000002bf7460ae00] max_analyze_duration 20000000 reached at 20000000 microseconds st:0
  
  [mpegts @ 000002bf7460ae00] start time for stream 2 is not set in estimate_timings_from_pts
  
  Input #0, mpegts, from 'TEMP1.ts':
  
    Duration: 01:51:54.48, start: 1.440000, bitrate: 7992 kb/s
  
    Program 1
  
      Metadata:
  
        service_name    : Service01
  
        service_provider: FFmpeg
  
      Stream #0:0[0x100]: Video: h264 (High), 1 reference frame ([27][0][0][0] / 0x001B), yuv420p(progressive, left), 1920x1080 (1920x1088) [SAR 1:1 DAR 16:9], 25 fps, 25 tbr, 90k tbn, 50 tbc
  
      Stream #0:1[0x101](eng): Audio: aac (LC) ([15][0][0][0] / 0x000F), 48000 Hz, stereo, fltp, 254 kb/s
  
      Stream #0:2[0x102](dan): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
  
  [vobsub @ 000002bf75d65940] IDX/SUB: out.fi.idx -> out.fi.sub
  
  Input #1, vobsub, from 'out.fi.idx':
  
    Duration: N/A, bitrate: N/A
  
      Stream #1:0[0x0](fi): Subtitle: dvd_subtitle, 1920x1080 (default)
  
  File 'TEMP2.ts' already exists. Overwrite? [y/N] y
  
  [mpegts @ 000002bf74634f00] service 1 using PCR in pid=256, pcr_period=20ms
  
  [mpegts @ 000002bf74634f00] muxrate 8002000, sdt every 500 ms, pat/pmt every 100 ms
  
  Output #0, mpegts, to 'TEMP2.ts':
  
    Metadata:
  
      encoder         : Lavf58.35.100
  
      Stream #0:0: Video: h264 (High), 1 reference frame ([27][0][0][0] / 0x001B), yuv420p(progressive, left), 1920x1080 (0x0) [SAR 1:1 DAR 16:9], q=2-31, 25 fps, 25 tbr, 90k tbn, 90k tbc
  
      Stream #0:1(eng): Audio: aac (LC) ([15][0][0][0] / 0x000F), 48000 Hz, stereo, fltp, 254 kb/s
  
      Stream #0:2(dan): Subtitle: dvb_subtitle ([6][0][0][0] / 0x0006)
  
      Stream #0:3(fin): Subtitle: dvb_subtitle (dvbsub), 1920x1080 (default)
  
      Metadata:
  
        encoder         : Lavc58.62.100 dvbsub
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (copy)
  
    Stream #0:1 -> #0:1 (copy)
  
    Stream #0:2 -> #0:2 (copy)
  
    Stream #1:0 -> #0:3 (dvd_subtitle (dvdsub) -> dvb_subtitle (dvbsub))
  
  Press [q] to stop, [?] for help
  
  No more output streams to write to, finishing.me=01:51:46.51 bitrate=7990.2kbits/s speed=18.2x
  
  frame=167862 fps=456 q=-1.0 Lsize= 6558750kB time=01:51:54.47 bitrate=8002.0kbits/s speed=18.2x
  
  video:5738580kB audio:212004kB subtitle:10989kB other streams:0kB global headers:0kB muxing overhead: 10.017118%
  
  Input file #0 (TEMP1.ts):
  
    Input stream #0:0 (video): 167862 packets read (5876305439 bytes);
  
    Input stream #0:1 (audio): 314742 packets read (217092443 bytes);
  
    Input stream #0:2 (subtitle): 1740 packets read (5578504 bytes);
  
    Total: 484344 packets (6098976386 bytes) demuxed
  
  Input file #1 (out.fi.idx):
  
    Input stream #1:0 (subtitle): 855 packets read (6881158 bytes); 854 frames decoded;
  
    Total: 855 packets (6881158 bytes) demuxed
  
  Output file #0 (TEMP2.ts):
  
    Output stream #0:0 (video): 167862 packets muxed (5876305439 bytes);
  
    Output stream #0:1 (audio): 314742 packets muxed (217092443 bytes);
  
    Output stream #0:2 (subtitle): 1740 packets muxed (5578504 bytes);
  
    Output stream #0:3 (subtitle): 1708 frames encoded; 1708 packets muxed (5673736 bytes);
  
    Total: 486052 packets (6104650122 bytes) muxed
  
  [AVIOContext @ 000002bf7463c280] Statistics: 0 seeks, 25621 writeouts
  
  [AVIOContext @ 000002bf74613f80] Statistics: 6728300448 bytes read, 2 seeks
  
  [AVIOContext @ 000002bf75efde40] Statistics: 7852032 bytes read, 0 seeks
  
  [AVIOContext @ 000002bf74f30e40] Statistics: 40275 bytes read, 0 seeks

  So i have video/audio and two dvb tracks inside one TS.

  Now im adding third subtitle and i get this warning :

  frame=32351 fps=869 q=-1.0 size= 1255936kB time=00:21:33.99 bitrate=7951.1kbits/s speed=34.8x
  
  [mpegts @ 0x2994600] Non-monotonous DTS in output stream 0:3; previous: 116868609, current: 116868607; changing to 116868610. This may result in incorrect timestamps in the output file.
  
  frame=33416 fps=886 q=-1.0 size= 1297408kB time=00:22:16.56 bitrate=7952.0kbits/s speed=35.4x

  By this warning it seems that something is not right with DVB (0:3) Finish track.
  Playing back the file with VLC i can see that Finish subtitles (between 21:33 and 22:16) are not showing up.

  Dumping packets from 0:3 track i see that 4 packets have the same DTS/PTS values.

  #stream#, dts,        pts, duration,     size, hash
  
  0,  111258000,  111258000,        0,     6979, 6215f060, S=1,        1, 67d220bc
  
  0,  111689100,  111689100,        0,       25, e60e94fa, S=1,        1, 67d220bc
  
  0,  113230800,  113230800,        0,     4423, f811be62, S=1,        1, 67d220bc
  
  0,  113561550,  113561550,        0,       25, 6db8d81f, S=1,        1, 67d220bc
  
  0,  113788800,  113788800,        0,     8247, 8b494779, S=1,        1, 67d220bc
  
  0,  114191190,  114191190,        0,       25, a205e183, S=1,        1, 67d220bc
  
  
  
  0,  116868607,  116868607,        0,     1119, e5d11188, S=1,        1, 67d220bc
  
  0,  116868607,  116868607,        0,       28, 11e09861, S=1,        1, 67d220bc
  
  0,  116868607,  116868607,        0,     1122, 8320a3f5, S=1,        1, 67d220bc
  
  0,  116868607,  116868607,        0,       28, e896fa37, S=1,        1, 67d220bc
  
  
  
  0,  117205200,  117205200,        0,     3583, 47b23f27, S=1,        1, 67d220bc
  
  0,  117611640,  117611640,        0,       25, e5ae32ed, S=1,        1, 67d220bc
  
  0,  117709200,  117709200,        0,     1119, 208b4f80, S=1,        1, 67d220bc
  
  0,  117830970,  117830970,        0,       25, 2a130b71, S=1,        1, 67d220bc
  
  0,  117846000,  117846000,        0,     3105, f179a4f0, S=1,        1, 67d220bc
  
  0,  118148040,  118148040,        0,       25, 29b3ad66, S=1,        1, 67d220bc
  
  0,  120146410,  120146410,        0,     1119, 714ef5d1, S=1,        1, 67d220bc
  
  0,  120146410,  120146410,        0,       28, bbad7fed, S=1,        1, 67d220bc
  
  0,  120708000,  120708000,        0,     2060, 703084f7, S=1,        1, 67d220bc
  
  0,  120855420,  120855420,        0,       25, 6db8d81f, S=1,        1, 67d220bc
  
  0,  121122000,  121122000,        0,     2184, 04a7be71, S=1,        1, 67d220bc
  
  0,  121395330,  121395330,        0,       25, a205e183, S=1,        1, 67d220bc

  Thes are the lines from SRT file where the warning happens :

  146
  
  00:21:34,840 --> 00:21:36,600
  
  Make!
  
  147
  
  00:21:42,280 --> 00:21:46,800
  
  Make! Hei, Make. Make!
  
  148
  
  00:21:47,880 --> 00:21:49,240
  
  Make!
  
  149
  
  00:21:49,400 --> 00:21:52,760
  
  Odota minua, Make!
  
  150
  
  00:22:00,400 --> 00:22:01,760
  
  Make!

  Sometimes muxing subtitles in different order (for example Finish before Danish and then Swedish) would solve this problem, but i would like to know why such things are happening.

• About the delay of Electron playing FFpmeg transcoded video

  26 juillet 2020, par Yohann

  In the Electron project, you need to try to play the video screen of the camera

  


  The camera is Haikang’s webcam

  


  Get real-time video stream through RTSP protocol

  


  rtsp://admin:admin@192.168.0.253/main/Channels/

  


  There will be a delay of about 3s when playing through the VLC player

  


  Then when used in the project, create a websocket service in Electron through the main process, decode the rtsp video through fluent-ffmpeg, and convert it to flv stream and push it to the rendering process

  


  import * as express from 'express'
import * as expressWebSocket from 'express-ws'
import ffmpeg from 'fluent-ffmpeg'
import webSocketStream from 'websocket-stream/stream'
const path = require('path')

let ffmpegPath
if (process.env.NODE_ENV === 'development') {
  ffmpegPath = path.join(__static, 'ffmpeg', 'bin', 'ffmpeg.exe')
} else {
  ffmpegPath = path.join(process.cwd(), 'ffmpeg', 'bin', 'ffmpeg.exe')
}
ffmpeg.setFfmpegPath(ffmpegPath)

// Start video transcoding service
function videoServer () {
  let app = express()
  app.use(express.static(__dirname))
  expressWebSocket(app, null, {
    perMessageDeflate: true
  })
  app.ws('/rtsp/', rtspRequestHandle)
  app.listen(8888)
  console.log('Create a monitoring service')
}

//RTSP transcoding method
function rtspRequestHandle (ws, req) {
  console.log('rtsp request handle')
  const stream = webSocketStream(ws, {
    binary: true,
    browserBufferTimeout: 1000000
  },
  {
    browserBufferTimeout: 1000000
  })
  let url = req.query.url
  console.log('rtsp url:', url)
  try {
    ffmpeg(url)
      .addInputOption('-rtsp_transport', 'tcp', '-buffer_size', '102400') // Here you can add some RTSP optimized parameters
      .outputOptions([
        '-fflags',
        'nobuffer',
        '-tune',
        'zerolatency'
      ])
      .on('start', function () {
        console.log(url, 'Stream started.')
      })
      .on('codecData', function () {
        console.log(url, 'Stream codecData.')
      })
      .on('error', function (err) {
        console.log(url, 'An error occured: ', err.message)
      })
      .on('end', function () {
        console.log(url, 'Stream end!')
      })
      .outputFormat('flv').videoCodec('copy').noAudio().pipe(stream)
  } catch (error) {
    console.log(error)
  }
}

export default videoServer


  


  The rendering process parses the video stream and plays the video through flv.js

  


  <template>&#xA;  <div class="video">&#xA;    <video class="video-box" ref="player"></video>&#xA;  </div>&#xA;</template>&#xA;&#xA;<code class="echappe-js">&lt;script&gt;&amp;#xA;  import flvjs from &amp;#x27;flv.js&amp;#x27;&amp;#xA;  export default {&amp;#xA;    name: &amp;#x27;videopage&amp;#x27;,&amp;#xA;    props: {&amp;#xA;      rtsp: String&amp;#xA;    },&amp;#xA;    data () {&amp;#xA;      return {&amp;#xA;        player: null&amp;#xA;      }&amp;#xA;    },&amp;#xA;    mounted () {&amp;#xA;      console.log(flvjs.isSupported())&amp;#xA;      if (flvjs.isSupported()) {&amp;#xA;        let video = this.$refs.player&amp;#xA;        console.log(video)&amp;#xA;        if (video) {&amp;#xA;          this.player = flvjs.createPlayer({&amp;#xA;            type: &amp;#x27;flv&amp;#x27;,&amp;#xA;            isLive: true,&amp;#xA;            url: &amp;#x27;ws://localhost:8888/rtsp/?url=&amp;#x27; &amp;#x2B; this.rtsp&amp;#xA;          }, {&amp;#xA;            enableStashBuffer: true&amp;#xA;          })&amp;#xA;          console.log(this.player)&amp;#xA;          this.player.attachMediaElement(video)&amp;#xA;          try {&amp;#xA;            this.player.load()&amp;#xA;            this.player.play()&amp;#xA;          } catch (error) {&amp;#xA;            console.log(error)&amp;#xA;          }&amp;#xA;        }&amp;#xA;      }&amp;#xA;    },&amp;#xA;    methods: {&amp;#xA;      getCurrentFrame () {&amp;#xA;        let video = this.$refs.player&amp;#xA;        let scale = 1&amp;#xA;        let canvas = document.createElement(&amp;#x27;canvas&amp;#x27;)&amp;#xA;        canvas.width = video.videoWidth * scale&amp;#xA;        canvas.height = video.videoHeight * scale&amp;#xA;        canvas.getContext(&amp;#x27;2d&amp;#x27;).drawImage(video, 0, 0, canvas.width, canvas.height)&amp;#xA;        return canvas.toDataURL(&amp;#x27;image/png&amp;#x27;)&amp;#xA;      }&amp;#xA;    },&amp;#xA;    beforeDestroy () {&amp;#xA;      this.player &amp;amp;&amp;amp; this.player.destory &amp;amp;&amp;amp; this.player.destory()&amp;#xA;    }&amp;#xA;  }&amp;#xA;&lt;/script&gt;&#xA;&#xA;&#xA;

  


  Then there will be a delay of about 3s when playing, and the delay will increase with the playing time

  


  There will be a delay of 10 minutes in the later period

  


  Is there any way to control this delay time

  


  Or is there any other decoding scheme that can be used ?

  


  Solutions that can support electron