Recherche avancée

Sur d’autres sites (2662)

• Is it advisable to use FFMpeg on my local server for video conversion ?

  7 juin 2019, par Yash Desai

  We are starting a video sharing website where users will be able to upload videos in their native formats. However, since video streaming on the web generally is in the FLV format, we need to convert the videos to FLV.

  Also, the site will be hosted on Amazon EC2 and storage using S3.

  Can i run FFMpeg on amazon EC2 ? Is this the best way to go ? Are there other alternatives to video encoding rather than doing conversion on our own server ? I also came across www.transloadit.com which seems to do the same but they are charging a bomb. Are there cheaper and more intelligent alternatives ?

  We are planning to make this website as one of top 10 biggest niche video streaming websites on the internet.

• FFMPEG - Output video not playing after effect

  11 mai 2021, par D.B.

  I am using FFMPEG for giving effects to video.

  


  I tried to give effect to video using below commands.

  


  ffmpeg -i Input1.mp4 -i Input2.mp4 -filter_complex xfade=transition=circleopen:duration=5:offset=0 OutPutVideo.mp4


  


  My command runs fine but output video is not running even in VLC media player.

  


  I got these commands from location :
https://ottverse.com/crossfade-between-videos-ffmpeg-xfade-filter/

  


  My videos are at location :
https://drive.google.com/drive/folders/1SLsrRUjyGH3eM7Oe3cptU_YPkga7EoCO?usp=sharing

  


  Please advise what is the issue ?

  


  Kind Regards,

  


• Strange results when converting rgb yuv using ffmpeg in python

  20 avril 2020, par user3569998

  I have JPG images and I am trying to do the following :

  



  

  1. resize the input and save the result into PNG

  


  2. convert PNG images to YUV (for instance yuv444p10le) in AVI container

  


  3. convert AVI back to PNG image.

  


  



  I am using python 3 and ffmpeg Linux version installed on Colab

  



  I printed the difference between resized_png and restored_png. I have values of 255 which doesn't make sense at all.

  



   [[[  0 255   1]
  [  0   0   1]
  [  0 255   1]
  ...
  [  1   0   0]
  [  1   0   0]
  [  2   0   1]]

 [[  0 255   1]
  [  0 255   1]
  [  1   0   0]
  ...
  [  1   0   0]
  [  1   0   0]
  [  1   0   0]]

 [[  0   0   1]
  [  0   0   1]
  [  0   0   1]
  ...
  [  1 255   0]
  [  1 255   1]
  [  1   0   0]]

 ...

 [[255   0   0]
  [255   0   1]
  [255   0   1]
  ...
  [  0   0   0]
  [  0   0   1]
  [255   0   1]]

 [[255   0   1]
  [255   0   1]
  [255   0   1]
  ...
  [  0   0   1]
  [  0   0   1]
  [  0   0   0]]

 [[255   0   1]
  [255   0   1]
  [255   0   1]
  ...
  [255   1   0]
  [255   0   0]
  [255   0   1]]]


  



  input_image = '/content/drive/My Drive/Colab Notebooks/adv_dnn/datasets/im2.jpg'
output_resized = '/content/drive/My Drive/Colab Notebooks/adv_dnn/datasets/im2.png'
folder_path = '/content/drive/My Drive/Colab Notebooks/adv_dnn/datasets/'

#do resize
# code = subprocess.call('cd /usr/bin/ffmpeg',  shell=True)
# print(code)
cmd_resize = ['ffmpeg', '-y', '-i',  input_image,'-vf', 'scale=224:224', output_resized]
process = subprocess.Popen(cmd_resize, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
out, err = process.communicate()
print(out)
print(err)

fmt = ['yuv420p', 'yuv420p10le', 'yuv444p', 'yuv444p10le']
for f in fmt:
  cmd2YUV = ['ffmpeg', '-y', '-i', output_resized, '-c:v', 'libx264', '-preset', 'placebo',\
             '-qp', '0', '-x264-params', "keyint=15:no-deblock=1", '-pix_fmt', f, \
             '-sws_flags', 'spline+accurate_rnd+full_chroma_int', \
             '-vf', "colorspace=bt709:iall=bt601-6-625:fast=1", '-color_range', '1', '-colorspace', '1', 
             '-color_primaries', '1', '-color_trc', '1', folder_path+'im2_'+f+'.avi']
  process = subprocess.Popen(cmd2YUV, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
  out, err = process.communicate()
  print(out)
  print(err)

  cmd2PNG = ['ffmpeg', '-y', '-i',  folder_path+'im2_'+f+'.avi', '-compression_level', '10', '-pred', 'mixed', \
             '-pix_fmt', 'rgb24', '-sws_flags', '+accurate_rnd+full_chroma_int', \
             folder_path+'im2_'+f+'.png']
  process = subprocess.Popen(cmd2PNG, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
  out, err = process.communicate()
  print(out)
  print(err)

im1=pilim.open('/content/drive/My Drive/Colab Notebooks/adv_dnn/datasets/im2.png')
im2=pilim.open('/content/drive/My Drive/Colab Notebooks/adv_dnn/datasets/im2_yuv444p10le.png')
im1_2d = np.asarray(im1)
im2_2d = np.asarray(im2)
print(im2_2d-im1_2d)