Recherche avancée

Sur d’autres sites (7403)

• Capture Video with opencv, save to ffmpeg pipe and live stream

  9 mars 2018, par Chris

  The goal is to stream an analysed live video over RTSP to some media server. To make the edits/analysis I use opencv, save the edited frames as JPEG in an FFMPEG image pipe and use the same FFMPEG to create a RTSP stream. Sorry if the terminology is not that accurate, I find it still quite confusing.

  I have the following code after quite some struggle :

  import cv2
  
  from subprocess import Popen, PIPE
  
  from PIL import Image
  
  
  
  # open pipe
  
  p = Popen('ffmpeg -y -f image2pipe -vcodec mjpeg -r 24 -i - -vcodec h264 -f rtsp -rtsp_transport tcp rtsp://localhost:8081/test.sdp', stdin=PIPE)
  
  
  
  video = cv2.VideoCapture(0)
  
  i = 0
  
  while video.isOpened():
  
      i=i+1
  
      ret, frame = video.read()
  
      if ret:
  
          #[...do some analysis stuff]
  
          im = Image.fromarray(frame)
  
          im.save(p.stdin, 'JPEG')
  
  
  
          """
  
          alternatively
  
          img_str = cv2.imencode('.jpg', frame)[1].tostring()
  
          p.stdin.write(img_str)
  
          """
  
  
  
      else:
  
          break
  
  
  
      print (i)
  
      if(i==1000):
  
          break
  
  
  
  
  
  p.stdin.close()
  
  p.wait()
  
  video.release()
  
  cv2.destroyAllWindows()
  
  print("done streaming video")

  This runs for 124 frames (i=124) then the loop hangs and I get some message from ffmpeg where I am not sure what it is about, however it does not look like an error :

  push frame
  
  122
  
  push frame
  
  123
  
  push frame
  
  124
  
  Input #0, image2pipe, from 'pipe:':
  
    Duration: N/A, bitrate: N/A
  
      Stream #0:0: Video: mjpeg, yuvj420p(pc, bt470bg/unknown/unknown), 640x480 [SAR 1:1 DAR 4:3], 24 fps, 24 tbr, 24 tbn, 24 tbc
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (mjpeg (native) -> h264 (libx264))
  
  [libx264 @ 000002076650d980] using SAR=1/1
  
  [libx264 @ 000002076650d980] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX
  
  [libx264 @ 000002076650d980] profile High, level 3.0
  
  [libx264 @ 000002076650d980] 264 - core 155 r2893 b00bcaf - H.264/MPEG-4 AVC codec - Copyleft 2003-2017 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=6 lookahead_threads=1 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=24 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
  
  push frame

  The webcam seems to continue running but no more frames are pushed into the pipe. It looks like some buffer is filled or something. If I write directly to a video file instead of rtsp, it works. If I open the rtsp stream simultaneously with ffplay, it also works (although with a 5 seconds lag).
  Anyone an idea where this is coming from and how to solve it ?

• ffmpeg low speed and wrong bitrate

  31 août 2018, par ngc3o34

  My goal is to compress video making it’s bitrate lower.
  For experiment I used following command

  ffmpeg -ss 00:00:00 -i V.mkv -to 00:10:00 -c:a copy -b:v 7500k -maxrate 7500k -minrate 5500k -bufsize 1M V7500.mkv

  1. What are possible reasons speed of conversion was close to 1.0x utilizing all 8 cores ?
  2. Bitrate of output video stream is about 4700 (5500 was set as minrate). Is there proper way to specify bitrate ?

  Stream mapping:
  
    Stream #0:0 -> #0:0 (h264 (native) -> h264 (libx264))
  
    Stream #0:1 -> #0:1 (copy)
  
  Press [q] to stop, [?] for help
  
  [libx264 @ 0x55e54ad189a0] using SAR=1/1
  
  [libx264 @ 0x55e54ad189a0] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX FMA3 BMI2 AVX2
  
  [libx264 @ 0x55e54ad189a0] profile High, level 4.0
  
  [libx264 @ 0x55e54ad189a0] 264 - core 152 r2854 e9a5903 - H.264/MPEG-4 AVC codec - Copyleft 2003-2017 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=12 lookahead_threads=2 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=23 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=cbr mbtree=1 bitrate=7500 ratetol=1.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 vbv_maxrate=7500 vbv_bufsize=1000 nal_hrd=none filler=0 ip_ratio=1.40 aq=1:1.00
  
  Output #0, matroska, to 'V7500.mkv':
  
    Metadata:
  
      title           : Valerian.and.the.City.of.a.Thousand.Planets.2017.1080p.BluRay.x264.DTS-HD.MA.7.1-FGT
  
      encoder         : Lavf57.83.100
  
      Chapter #0:0: start 0.000000, end 600.000000
  
      Metadata:
  
        title           : Chapter 01
  
      Stream #0:0(eng): Video: h264 (libx264) (H264 / 0x34363248), yuv420p, 1920x804 [SAR 1:1 DAR 160:67], q=-1--1, 7500 kb/s, 23.98 fps, 1k tbn, 23.98 tbc (default)
  
      Metadata:
  
        title           : Valerian.and.the.City.of.a.Thousand.Planets.2017.1080p.BluRay.x264.DTS-HD.MA.7.1-FGT
  
        DURATION        : 02:17:03.277000000
  
        encoder         : Lavc57.107.100 libx264
  
      Side data:
  
        cpb: bitrate max/min/avg: 7500000/0/7500000 buffer size: 1000000 vbv_delay: -1
  
      Stream #0:1(eng): Audio: aac (LC) ([255][0][0][0] / 0x00FF), 48000 Hz, stereo, fltp (default)
  
      Metadata:
  
        title           : Valerian.and.the.City.of.a.Thousand.Planets.2017.1080p.BluRay.x264.DTS-HD.MA.7.1-FGT
  
        ENCODER         : Lavc57.107.100 aac
  
        DURATION        : 02:17:03.317000000
  
  Past duration 0.994987 too large
  
  frame=14385 fps= 31 q=-1.0 Lsize=  363641kB time=00:09:59.99 bitrate=4964.9kbits/s speed=1.28x    

• Convert images into video

  24 septembre 2017, par Disha Shukla

  I have some images in the folder. I want to convert all images into videos. But the problem is that when the resolution is different it shows the error. I don’t want to stretch the images in the video. I just want to keep image size as it is. But it shows this error.

  My Command :

  ffmpeg -framerate 1/3 -pattern_type glob -i "images/*.jpg" -i trinity.MP3 -shortest -c:v libx264 -r 30 -vf "scale=iw*sar:ih , pad=max(iw\,ih*(16/9)):ow/(16/9):(ow-iw)/2:(oh-ih)/2" -aspect 16:9 -pix_fmt yuv420p out.mp4 -y

  Logs :

  [mjpeg @ 0x6b3820] Changeing bps to 8
  
  Input #0, image2, from 'images/*.jpg':
  
    Duration: 00:00:24.00, start: 0.000000, bitrate: N/A
  
      Stream #0:0: Video: mjpeg, yuvj420p(pc, bt470bg/unknown/unknown), 1280x720 [SAR 1:1 DAR 16:9], 0.33 fps, 0.33 tbr, 0.33 tbn, 0.33 tbc
  
  [mp3 @ 0x6b4ca0] Estimating duration from bitrate, this may be inaccurate
  
  Input #1, mp3, from 'trinity.MP3':
  
    Metadata:
  
      encoded_by      : Lavf52.9.0
  
    Duration: 00:09:02.35, start: 0.000000, bitrate: 128 kb/s
  
      Stream #1:0: Audio: mp3, 44100 Hz, stereo, s16p, 128 kb/s
  
  [swscaler @ 0x68a3c0] deprecated pixel format used, make sure you did set range correctly
  
  [libx264 @ 0x6bada0] using SAR=1/1
  
  [libx264 @ 0x6bada0] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX AVX2 FMA3 LZCNT BMI2
  
  [libx264 @ 0x6bada0] profile High, level 3.1
  
  [libx264 @ 0x6bada0] 264 - core 142 r2438 af8e768 - H.264/MPEG-4 AVC codec - Copyleft 2003-2014 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=3 lookahead_threads=1 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=25 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
  
  Output #0, mp4, to 'out.mp4':
  
    Metadata:
  
      encoder         : Lavf56.25.101
  
      Stream #0:0: Video: h264 (libx264) ([33][0][0][0] / 0x0021), yuv420p, 1280x720 [SAR 1:1 DAR 16:9], q=-1--1, 30 fps, 15360 tbn, 30 tbc
  
      Metadata:
  
        encoder         : Lavc56.26.100 libx264
  
      Stream #0:1: Audio: aac (libfaac) ([64][0][0][0] / 0x0040), 44100 Hz, stereo, s16, 128 kb/s
  
      Metadata:
  
        encoder         : Lavc56.26.100 libfaac
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (mjpeg (native) -> h264 (libx264))
  
    Stream #1:0 -> #0:1 (mp3 (native) -> aac (libfaac))
  
  Press [q] to stop, [?] for help
  
  Input stream #0:0 frame changed from size:1280x720 fmt:yuvj420p to size:596x913 fmt:yuvj420p
  
  [Parsed_pad_1 @ 0x697940] Input area 512:0:1108:913 not within the padded area 0:0:1622:912 or zero-sized
  
  [Parsed_pad_1 @ 0x697940] Failed to configure input pad on Parsed_pad_1

  Is there any way to keep resolution as it is and make a video of all images ?