Recherche avancée

Sur d’autres sites (8370)

• First input link main timebase (1/15360) do not match the corresponding second input link xfade timebase (1/90000)

  14 novembre 2023, par Rino

  coders

  


  I am new to ffmpeg and i am facing this error for more then 2 days.
This ffmpeg command :
!ffmpeg -i "{fromPath}" -i "{toPath}" -filter_complex "xfade=transition={chosenTransition}:duration={transitionDuration}:offset=-1.5" -c:v libx264 -crf 18 -preset veryfast "{outPath}"
I run that previous command 3 times in row where output of the previous command is input that is getting produced as 15360 timebase video i guess.

  


  is giving me this error : "First input link main timebase (1/15360) do not match the corresponding second input link xfade timebase (1/90000)"

  


  So when i merge two of my videos the output is video with (1/15360) timebase. Original videos has (1/90000) timebase.

  


  Pls help what i wana archive is basicly multiple videos inside a folder called "clips" merge together using a random transition from xfade/ffmpeg.

  


  On colab i tried all this ways :

  


  !rm -rf chunks
!mkdir chunks
time.sleep(1)
folderPath = '/content/clips'
videoOutput = '/content/test.mp4'
transitionDuration = 6

videoFiles = [f"{folderPath}/{f}" for f in os.listdir(folderPath) if f.endswith('.mp4')]
random.shuffle(videoFiles)
fromPath = videoFiles[0]
toPath = ''
outPath = ''
transitions = ['fade', 'wipeup', 'slidedown', 'circleclose', 'radial' , 'hrslice', 'hblur', 'smoothleft', 'smoothright']

for videoIndex, videoName in enumerate(videoFiles):
    if videoIndex == 0:
        continue
    toPath = videoName
    chosenTransition = random.choice(transitions)
    print('chosenTransition', chosenTransition)
    clipDuration = int((imageio.get_reader(fromPath)).get_meta_data()['duration']) - transitionDuration
    # !echo -i /content/clips/clip0.mp4 -i /content/clips/clip1.mp4 -filter_complex xfade=transition={chosenTransition}:duration=3:offset={clipDuration} dissolveVideo.mp4

    if videoIndex == len(videoFiles) - 1:
        outPath = videoOutput
    else:
        outPath = f'/content/chunks/chunk{videoIndex}.mp4'
    time.sleep(2)
    
    # !echo -i "{fromPath}" -i "{toPath}" -filter_complex "[0:v]xfade=transition={chosenTransition}:duration=3:offset={clipDuration},settb=1/19000[v0];[0:a][1:a]acrossfade=d=3[aout]" -map "[v0]" -map "[aout]" -c:v libx264 -crf 18 -preset ultrafast "{outPath}"
    !ffmpeg -i "{fromPath}" -i "{toPath}" -filter_complex "xfade=transition={chosenTransition}:duration={transitionDuration}:offset=-1.5,settb=1/15360" -c:v libx264 -crf 18 -preset veryfast "{outPath}"
    # !ffmpeg -i "{fromPath}" -i "{toPath}" -filter_complex "[0:v]settb=1/15360,setpts=PTS-STARTPTS[v0];[1:v]settb=1/15360,setpts=PTS-STARTPTS[v1];[v0][v1]xfade=transition={chosenTransition}:duration={transitionDuration}:offset=-1.5[vout]" -map "[vout]" -c:v libx264 -crf 18 -preset ultrafast "{outPath}"
    # !ffmpeg -i "{fromPath}" -i "{toPath}" -filter_complex "[0]settb=AVTB [0v]; [1]settb=AVTB [1v]; [0v] [1v]xfade=transition={chosenTransition}:duration={transitionDuration}:offset=1" -c:v libx264 -crf 18 -preset ultrafast "{outPath}"
    # !ffmpeg -i "{fromPath}" -i "{toPath}" -filter_complex "[0:v]xfade=transition={chosenTransition}:duration=3:offset={clipDuration},settb=1/15360[v0];[0:a][1:a]acrossfade=d=3[aout]" -map "[v0]" -map "[aout]" -c:v libx264 -crf 18 -preset ultrafast "{outPath}"
    # !ffmpeg -y -progress .progressinfo.dat -i "{fromPath}" -i "{toPath}" -filter_complex "[0:v]settb=AVTB,setpts=PTS-STARTPTS[v0];[1:v]settb=AVTB,setpts=PTS-STARTPTS[v1];[v0][v1]xfade=transition={chosenTransition}:duration={transitionDuration}:offset={clipDuration}[vout];[0:a]asettb=AVTB,asetpts=PTS-STARTPTS[a0];[1:a]asettb=AVTB,asetpts=PTS-STARTPTS[a1];[a0][a1]acrossfade=d=3[aout]" -map "[vout]" -map "[aout]" -c:v libx264 -crf 18 -preset ultrafast "{outPath}"



  


• openCV image Stitching wide angle 160 degrees

  4 janvier 2020, par a.masri

  I’m trying to Stitching image wide angle 160.5 degree but the result is not a good

  i’m using OpenCV 4 and ffmpeg to get frames from video

  ffmpeg command to get 15 frame per sec :

  ffmpeg -i first.mp4 -vf fps=15  preview%05d.jpg

  OpenCV Stitching code

  import cv2
  
  import numpy as np
  
  
  
  images = []
  
  for i in range(70):
  
      name = ('preview%05d.jpg' % (i+1))
  
      print(name)
  
      images.append(cv2.imread(name , cv2.IMREAD_COLOR))
  
  
  
  
  
  print("start ")
  
  stitcher = cv2.Stitcher_create()
  
  ret, pano = stitcher.stitch(images)
  
  
  
  if ret == cv2.STITCHER_OK:
  
      cv2.imshow('panorama', pano)
  
      cv2.waitKey()
  
      cv2.destroyAllWindows()
  
  else:
  
      print(cv2.STITCHER_ERR_NEED_MORE_IMGS)
  
      print(cv2.STITCHER_ERR_HOMOGRAPHY_EST_FAIL)
  
      print(cv2.STITCHER_ERR_CAMERA_PARAMS_ADJUST_FAIL)
  
      print(ret)
  
      print('Error during stiching')

  actual result :

  

  expected result :

  

• Handbreak ffmpeg encode vs decode speed (x265)

  10 février 2023, par Demian

  I'm encoding videos in x265 to save same storage (Gopro, DJI, etc.). This is working well for me with Handbreak and I got some good CRF values for my use cases.

  


  My question is about the encoding preset. The slower it is the better Qualitiy per Bitrate you get, but the encoding time increases.

  


  But how the decoding time. Does it also consume more CPU power to watch (decode) a video which was created with a slow prestet then a fast preset ?

  


  Thank you !