Recherche avancée

Sur d’autres sites (7165)

• dnn : add DCO_RGB color order to enum DNNColorOrder

  6 mai 2021, par Ting Fu

  dnn : add DCO_RGB color order to enum DNNColorOrder
  Adding DCO_RGB color order to DNNColorOrder, since tensorflow model
  
  needs this kind of color oder as input.
  Signed-off-by : Ting Fu <ting.fu@intel.com>
  

  • [DH] libavfilter/dnn/dnn_backend_tf.c
  • [DH] libavfilter/dnn/dnn_io_proc.c
  • [DH] libavfilter/dnn_interface.h

• cv2/ffmpeg "grabFrame packet read max attempts exceeded" error after exactly reading certain number of frames

  26 juin, par banjaxing

  I am using OpenCV to extract frames from videos, run a segmentation AI model, and save the frames and masks to a folder. When I run my code to extract the frame from I encounter the error "grabFrame packet read max attempts exceeded" after processing a certain number of frames. This issue occurs consistently for the same videos across multiple environments.

  


  Error message :

  


  [ WARN:0@379.898] global cap_ffmpeg_impl.hpp:1541 grabFrame packet read max attempts exceeded, if your video have multiple streams (video, audio) try to increase attempt limit by setting environment variable OPENCV_FFMPEG_READ_ATTEMPTS (current value is 10000)


  


  Minimum Reproducible Example

  


  import os
import cv2

videofilename = "test.mp4"
capture = cv2.VideoCapture(videofilename)
frameNum = 0

createfolder = os.getcwd() + '/' + videofilename.split(".")[0] + '/'
if not os.path.exists(createfolder):
    os.makedirs(createfolder)
    os.makedirs(createfolder + "/frames/")

while True:
    success, frame = capture.read()
    if success is False:
        break
    frameNum += 1
    framedownloadname = videofilename.split(".")[0] + '-fr' + str(frameNum) + '.jpg'
    framedownloadloc = createfolder + '/frames/' + framedownloadname
    print(framedownloadloc)
    cv2.imwrite(framedownloadloc, frame)
    img = cv2.imread(framedownloadloc)
    img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

capture.release()


  


  As suggested in error, I increased the OPENCV_FFMPEG_READ_ATTEMPTS env variable up to 10000. However, this seems to have little to no effect on the number of frames before the error appears.

  


• Transcode video using celery and ffmpeg in django

  28 octobre 2015, par Robin

  I would like to transcode user uploaded videos using celery. I think first I should upload the video, and spawn a celery task for transcoding.

  Maybe something like this in the tasks.py :

  subprocess.call('ffmpeg -i path/.../original path/.../output')

  Just completed First steps with celery, so confused how to do so in the views.py and tasks.py. Also is it a good solution ? I would really appreciate your help and advice. Thank you.

  models.py :

  class Video(models.Model):
  
      user = models.ForeignKey(User)
  
      title = models.CharField(max_length=100)
  
      original = models.FileField(upload_to=get_upload_file_name)
  
      mp4_480 = models.FileField(upload_to=get_upload_file_name, blank=True, null=True)
  
      mp4_720 = models.FileField(upload_to=get_upload_file_name, blank=True, null=True)
  
      privacy = models.CharField(max_length=1,choices=PRIVACY, default='F')
  
      pub_date = models.DateTimeField(auto_now_add=True, auto_now=False)

  my incomplete views.py :

  @login_required
  
  def upload_video(request):
  
      if request.method == 'POST':
  
          form = VideoForm(request.POST, request.FILES)
  
          if form.is_valid():
  
              if form.cleaned_data:
  
                  user = request.user
  
                  #
  
                  #
  
                  # No IDEA WHAT TO DO NEXT
  
                  #
  
                  #
  
                  return HttpResponseRedirect('/')
  
  
  
      else:
  
          form = VideoForm()
  
          return render(request, 'upload_video.html', {
  
              'form':form
  
              })