Recherche avancée

Sur d’autres sites (6944)

• h264 : eliminate low_delay

  11 avril 2016, par Anton Khirnov

  h264 : eliminate low_delay
  It is always unconditionally initialized in decode_postinit() and then
  
  immediately used in one place further below. All the other places where
  
  it is accessed are just useless fluff.
  

  • [DBH] libavcodec/h264.c
  • [DBH] libavcodec/h264.h
  • [DBH] libavcodec/h264_slice.c

• Python gets stuck at pipe.stdin.write(image.tostring())

  10 mars 2020, par Vandana Rajan

  I am reading each frame of video and adding time stamp to it as given below.

  command = ['ffmpeg',
  
              '-y', # (optional) overwrite output file if it exists
  
              '-f', 'rawvideo', #Input is raw video
  
              '-pix_fmt', 'bgr24', #Raw video format
  
              '-s', str(int(width)) + 'x' + str(int(height)), # size of one frame
  
              '-i', '-', # The input comes from a pipe
  
              '-an', # Tells FFMPEG not to expect any audio
  
              '-vcodec', 'mpeg4',
  
              '-b:v', '10M', #Sets a maximum bit rate
  
              Output_name]
  
      #Open the pipe
  
      pipe = sp.Popen(command, stdin=sp.PIPE, stderr=sp.PIPE)
  
  
  
      print('Processing....')
  
      print(' ')
  
      #Reads through each frame, calculates the timestamp, places it on the frame and exports the frame to the output video.
  
      #import pdb
  
      #pdb.set_trace()
  
      while current_frame < total_frames:
  
          success, image = video.read()
  
          if success:
  
              elapsed_time = video.get(cv2.CAP_PROP_POS_MSEC)
  
              current_frame = video.get(cv2.CAP_PROP_POS_FRAMES)
  
              timestamp = initial + dt.timedelta(microseconds = elapsed_time*1000)
  
              cv2.putText(image, 'Date: ' + str(timestamp)[0:10], (50,int(height-150)), cv2.FONT_HERSHEY_COMPLEX_SMALL, 2, (255, 255, 255), 3)
  
              cv2.putText(image, 'Time: ' + str(timestamp)[11:-4], (50,int(height-100)), cv2.FONT_HERSHEY_COMPLEX_SMALL, 2, (255, 255, 255), 3)
  
              pipe.stdin.write(image.tostring())
  
              print('frame number',current_frame)
  
          else:
  
              print('video reader fail')    
  
      video.release()
  
      pipe.stdin.close()
  
      pipe.stderr.close()

  However, after around 18k frames, Python gets stuck at ’pipe.stdin.write(image.tostring())’. It does not produce any error, but simply hangs. How to resolve this issue ?

  Thanks in advance.

• Last subprocess call not working in concatenation code with FFMPEG. How should I go about fixing this ?

  28 novembre 2019, par S1mple

  clips = []
  
  
  
  #generates a list of mp4 files in a folder
  
  def clipFinder(CurrentDir, fileType):
  
      clips.clear()
  
      for r,d,f in os.walk(CurrentDir):
  
          for file in f:
  
              if fileType in file:
  
                  clips.append(r+file)
  
      random.shuffle(clips)
  
  
  
  #removes all files that have the string 'vod' in them as they cause problems during concatenation
  
  def removeVods(r):
  
      for f in clips:
  
          if 'vod' in clips:
  
              os.remove(r+f)
  
  
  
  #generates a string using the clips list to insert into the ffmpeg command
  
  def clipString():
  
      string = 'intermediate'
  
      clipList = []
  
      clipNum = 1
  
      for f in clips:
  
          clipList.append(string+str(clipNum)+'.ts'+'|')
  
          clipNum+=1
  
      string1 = ''.join(clipList)
  
      string2 = string1[0:len(string1)-1]
  
      return string2
  
  
  
  #concatenates the mp4 files in the clipString
  
  def concatFiles():
  
      clipFinder('***', '.mp4')
  
      removeVods('***')
  
      i = 0
  
      intermediates = []
  
      for f in clips:
  
          subprocess.call(['***', '-i', clips[i], '-c', 'copy', '-bsf:v', 'h264_mp4toannexb', '-f', 'mpegts', 'intermediate'+ str(i+1) +'.ts'])
  
          i += 1 
  
      clipsLength = len(clips)
  
      subprocess.call['***', '-i', '"concat:' + clipString() + '"', '-c', 'copy', '-bsf:a 
  
      aac_adtstoasc', 'output.mp4']

  I am trying to make a clip concatenator, but the last subprocess call won’t run and gives me no error. When I run the script the first FFmpeg call works fine and gives me my intermediate mp4 files, however, the second command, which works when I run it in terminal, does not work when I run it from python using subprocess.call.

  Problematic code :

  subprocess.call(['***', '-i', '"concat:' + clipString() + '"', '-c', 'copy', '-bsf:a aac_adtstoasc', 'output.mp4'])

  all places with * were paths such as : /davidscomputer/bin/ffmpeg/