Recherche avancée

Sur d’autres sites (2945)

• Something wrong with recording video and merging audio and video by using FFmpeg

  16 mars 2014, par VaFancy

  At the beginning, I have to admit that I am a newbie in FFmpeg and also raspbian. Unfortunately, I need to solve this problem as soon as possible.

  I tried to record videos on my Raspberry PI by using PIcamera and USB webcam separately. When I used command :

  ffmpeg -t 10 -f video4linux2 -r 25 -i /dev/video0 out.h264

  However, something was very wrong because there was nothing saved in that video file. I copy the log.

  [video4linux2,v4l2 @ 0x267ad80] The driver changed the time per frame from 1/25 to 1/30
  
  Input #0, video4linux2,v4l2, from '/dev/video0':
  
    Duration: N/A, start: 37562.908020, bitrate: 147456 kb/s
  
      Stream #0:0: Video: rawvideo (YUY2 / 0x32595559), yuyv422, 640x480, 147456 kb/s, 30 fps, 30 tbr, 1000k tbn, 1000k tbc
  
  File 'out.h264' already exists. Overwrite ? [y/N] y
  
  No pixel format specified, yuv422p for H.264 encoding chosen.
  
  Use -pix_fmt yuv420p for compatibility with outdated media players.
  
  [libx264 @ 0x267ca60] using cpu capabilities: ARMv6
  
  [libx264 @ 0x267ca60] profile High 4:2:2, level 3.0, 4:2:2 8-bit
  
  Output #0, h264, to 'out.h264':
  
    Metadata:
  
      encoder         : Lavf55.34.101
  
      Stream #0:0: Video: h264 (libx264), yuv422p, 640x480, q=-1--1, 90k tbn, 25 tbc
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (rawvideo -> libx264)
  
  Press [q] to stop, [?] for help
  
  DTS 140730117085, next:18648459 st:0 invalid dropping00.00 bitrate=N/A    
  
  PTS 140730117085, next:18648459 invalid dropping st:0
  
  DTS 140740452429, next:6538221 st:0 invalid dropping:00.00 bitrate=N/A    
  
  PTS 140740452429, next:6538221 invalid dropping st:0
  
  frame=   44 fps=1.1 q=28.0 size=      18kB time=00:00:00.04 bitrate=3712.6kbits/
  
  frame=   45 fps=0.8 q=28.0 size=      22kB time=00:00:00.08 bitrate=2262.8kbits/
  
  frame=   46 fps=0.7 q=28.0 size=      27kB time=00:00:00.12 bitrate=1813.4kbits/
  
  frame=   47 fps=0.6 q=28.0 size=      30kB time=00:00:00.16 bitrate=1559.2kbits/
  
  frame=   48 fps=0.5 q=28.0 size=      34kB time=00:00:00.20 bitrate=1397.7kbits/
  
  frame=   49 fps=0.5 q=28.0 size=      39kB time=00:00:00.24 bitrate=1324.3kbits/
  
  frame=   50 fps=0.4 q=28.0 size=      42kB time=00:00:00.28 bitrate=1241.7kbits/ 

  As you can see, the size of each frame was only few kb and fps was so low. Thus each frame was taken by using a long time(4s in this case). Then I test my USB webcam by using command :

  ffplay -f video4linux2 /dev/video0

  The video stream which displayed on the screen was so unnatural. It looked like there was such a long delay between each frame. So the video was hardly moved. The log I have copied :

  pi@raspberrypi ~ $ ffplay -f video4linux2 /dev/video0
  
  avplay version 0.8.6-6:0.8.6-1+rpi1, Copyright (c) 2003-2013 the Libav developers
  
    built on Mar 31 2013 13:58:10 with gcc 4.6.3
  
  [video4linux2 @ 0x1ad83c0] Estimating duration from bitrate, this may be inaccurate
  
  Input #0, video4linux2, from '/dev/video0':
  
    Duration: N/A, start: 36107.154598, bitrate: 147456 kb/s
  
      Stream #0.0: Video: rawvideo, yuyv422, 640x480, 147456 kb/s, 30 tbr, 1000k tbn, 30 tbc
  
  [avsink @ 0x1ad9960] auto-inserting filter 'auto-inserted scaler 0' between the filter 'src' and the filter 'out'
  
  [scale @ 0x1ad9d20] w:640 h:480 fmt:yuyv422 -> w:640 h:480 fmt:yuv420p flags:0x4
  
  36163.33 A-V:  0.000 s:0.0 aq=    0KB vq= 3000KB sq=    0B f=8/8 

  The final problem is that no sound was added in video file when I tried to add some in. I used the command :

  avconv -i /home/pi/Desktop/test.mp4 -i /home/pi/Desktop/test.mp3 -map 0:0 -vcodec copy -map 1:0 -acodec copy output.h264

  I have checked separately that the video and audio were both able to play. However the merged file only had video but no audio. The log is shown below, it seems that the process was succeed.

  pi@raspberrypi ~ $ avconv -i /home/pi/Desktop/test.mp4 -i /home/pi/Desktop/test.mp3 -map 0:0 -vcodec copy -map 1:0 -acodec copy output.h264
  
  avconv version 0.8.6-6:0.8.6-1+rpi1, Copyright (c) 2000-2013 the Libav developers
  
    built on Mar 31 2013 13:58:10 with gcc 4.6.3
  
  [h264 @ 0x1aa6680] max_analyze_duration reached
  
  [h264 @ 0x1aa6680] Estimating duration from bitrate, this may be inaccurate
  
  Input #0, h264, from '/home/pi/Desktop/test.mp4':
  
    Duration: N/A, bitrate: N/A
  
      Stream #0.0: Video: h264 (High), yuv420p, 640x480, 25 fps, 25 tbr, 1200k tbn, 2400k tbc
  
  [mp3 @ 0x1aaa120] Estimating duration from bitrate, this may be inaccurate
  
  Input #1, mp3, from '/home/pi/Desktop/test.mp3':
  
    Metadata:
  
      title           : We Are Legend (Original Mix)
  
      artist          : Dimitri Vegas & Like Mike & Steve Aoki
  
      album           : SPINNIN RECORDS
  
      encoded_by      : iTunes 11.1.5
  
      TBP             : 128
  
      genre           : Progressive House
  
      date            : 2013
  
    Duration: 00:00:11.14, start: 0.000000, bitrate: 191 kb/s
  
      Stream #1.0: Audio: mp3, 48000 Hz, stereo, s16, 192 kb/s
  
  File 'output.h264' already exists. Overwrite ? [y/N] y
  
  Output #0, h264, to 'output.h264':
  
    Metadata:
  
      encoder         : Lavf53.21.1
  
      Stream #0.0: Video: libx264, yuv420p, 640x480, q=2-31, 90k tbn, 1200k tbc
  
      Stream #0.1: Audio: libmp3lame, 48000 Hz, stereo, 192 kb/s
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (copy)
  
    Stream #1:0 -> #0:1 (copy)
  
  Press ctrl-c to stop encoding
  
  frame=  144 fps=  0 q=-1.0 Lsize=    1684kB time=5.06 bitrate=2724.7kbits/s    
  
  video:1566kB audio:119kB global headers:0kB muxing overhead 0.000000%

  Could somebody helps me, please. And sorry for my bad english.

  BTW : When I finally captured 200 frames, it turns out all of them were green.

• /dev/nvidia0 opened multiple times while transcoding with ffmpeg

  22 novembre 2017, par sandeep

  While transcoding a video with ffmpeg hardware acceleration,the NVIDIA device getting opened multiple times.

  I am unable to understand why the device got opened those many times.

• Android built in Camera encoder VS FFMPEG --- Speed

  6 décembre 2017, par user3903275

  How does recording a 1080P, H264 encoded video in android camera application is realtime fast but encoding a video in android using FFMPEG is slow at the same video size ?

  I know FFMPEG is a software level encoder and it wont support any hardware features.

  I know camera applications directly get buffer data from camera driver.

  But actually where the difference happens ??

  Why camera application is Realtime fast ???

  Does it use GPU and OpenGL features of the phone to encode the video so that its so realtime fast ??

  Because both Camera Application and FFMPEG runs on same mobile but still camera encodes H264 realtime ???