Recherche avancée

Sur d’autres sites (11437)

• Extract different types of frames using FFMPEG [closed]

  24 juillet 2024, par Çağatay Alptekin

  How should I extract I, P, and B frames from video using the FFMPEG library ? the first stage would be using VideoCapture from OpenCV. I guess. My end goal is to do this with C# using FFMPEG.Autogen.

  


• Saving an animation using ffmpeg and matplotlib on anaconda3

  21 juin 2016, par Varsha Dyavaiah

  I am trying to create videos of NBA Action with Sportsvu data.

  I was following the steps given in this blog by Dan Vatterott :

  http://www.danvatterott.com/blog/2016/06/16/creating-videos-of-nba-action-with-sportsvu-data/?utm_campaign=Data%2BElixir&utm_medium=email&utm_source=Data_Elixir_84

  I am trying to create a animation and save it using ffmpeg and matplotlib.
  The code snippet is attached below.

   import matplotlib.animation as animation
  
   plt.rcParams['animation.ffmpeg_path'] = '/home/anaconda3/pkgs/ffmpeg-2.1.0-1/bin'
  
  
  
   fig = plt.figure(figsize=(15,7.5)) #create figure object
  
   ax = plt.gca() #create axis object
  
  
  
   draw_court([0,100,0,50]) #draw the court
  
   player_text = list(range(10)) #create player text vector
  
   player_circ = list(range(10)) #create player circle vector
  
   ball_circ = plt.Circle((0,0), 1.1, color=[1, 0.4, 0]) #create circle object for bal
  
   for i in list(range(10)): #create circle object and text object for each player
  
       col=['w','k'] if i<5 else ['k','w'] #color scheme
  
       player_circ[i] = plt.Circle((0,0), 2.2, facecolor=col[0],edgecolor='k') #player circle
  
       player_text[i] =   ax.text(0,0,'',color=col[1],ha='center',va='center') #player jersey # (text)
  
  
  
   ani = animation.FuncAnimation(fig, animate,  frames=np.arange(0,np.size(ball_xy,0)), init_func=init, blit=True, interval=5, repeat=False,\
  
                           save_count=0) #function for making video
  
  
  
   FFwriter = animation.FFMpegWriter()
  
   ani.save('Event_%d.mp4' % (search_id),dpi=100,writer = FFwriter,fps=25) #function for saving video
  
   plt.close('all') #close the plot

  When I try to save the animation ’ani’ , I get Errno 13 (Permission denied).

   ---------------------------------------------------------------------------
  
   PermissionError                           Traceback (most recent call last)
  
   in <module>()
  
       17 
  
       18 FFwriter = animation.FFMpegWriter()
  
  ---> 19 ani.save('Event_%d.mp4' % (search_id),dpi=100,writer = FFwriter,fps=25) #function for saving video
  
   20 plt.close('all') #close the plot
  
  
  
  /home/anaconda3/lib/python3.5/site-packages/matplotlib/animation.py in save(self, filename, writer, fps, dpi, codec, bitrate, extra_args, metadata, extra_anim, savefig_kwargs)
  
      799         # since GUI widgets are gone. Either need to remove extra code to
  
      800         # allow for this non-existant use case or find a way to make it work.
  
  --> 801         with writer.saving(self._fig, filename, dpi):
  
      802             for anim in all_anim:
  
      803                 # Clear the initial frame
  
  
  
  /home/anaconda3/lib/python3.5/contextlib.py in __enter__(self)
  
       57     def __enter__(self):
  
       58         try:
  
  ---> 59             return next(self.gen)
  
       60         except StopIteration:
  
       61             raise RuntimeError("generator didn't yield") from None
  
  
  
  /home/anaconda3/lib/python3.5/site-packages/matplotlib/animation.py in saving(self, *args)
  
      192         '''
  
      193         # This particular sequence is what contextlib.contextmanager wants
  
  --> 194         self.setup(*args)
  
      195         yield
  
      196         self.finish()
  
  
  
  /home/anaconda3/lib/python3.5/site-packages/matplotlib/animation.py in setup(self, fig, outfile, dpi, *args)
  
      182         # Run here so that grab_frame() can write the data to a pipe. This
  
      183         # eliminates the need for temp files.
  
  --> 184         self._run()
  
      185 
  
      186     @contextlib.contextmanager
  
  
  
  /home/anaconda3/lib/python3.5/site-packages/matplotlib/animation.py in _run(self)
  
      210                                       stdout=output, stderr=output,
  
      211                                       stdin=subprocess.PIPE,
  
  --> 212                                       creationflags=subprocess_creation_flags)
  
      213 
  
      214     def finish(self):
  
  
  
  /home/anaconda3/lib/python3.5/subprocess.py in __init__(self, args, bufsize, executable, stdin, stdout, stderr, preexec_fn, close_fds, shell, cwd, env, universal_newlines, startupinfo, creationflags, restore_signals, start_new_session, pass_fds)
  
      948                                 c2pread, c2pwrite,
  
      949                                 errread, errwrite,
  
  --> 950                                 restore_signals, start_new_session)
  
      951         except:
  
      952             # Cleanup if the child failed starting.
  
  
  
  /home/anaconda3/lib/python3.5/subprocess.py in _execute_child(self, args, executable, preexec_fn, close_fds, pass_fds, cwd, env, startupinfo, creationflags, shell, p2cread, p2cwrite, c2pread, c2pwrite, errread, errwrite, restore_signals, start_new_session)
  
     1542                             else:
  
     1543                                 err_msg += ': ' + repr(orig_executable)
  
  -> 1544                     raise child_exception_type(errno_num, err_msg)
  
     1545                 raise child_exception_type(err_msg)
  
     1546 
  
  
  
  PermissionError: [Errno 13] Permission denied
  
  </module>

  Can someone help me ? Thanks in advance.

• Subprocess overhead vs library function ?

  30 décembre 2024, par Liran

  I'm writing an application, and I want to perform hardware encoding (av1. My hardware is capable of that) of frames.

  


  I seem to have two options - either call ffmpeg as a child process, or dynamically link ffmpeg/gstreamer and call the library functions directly. The child process option will be significantly easier.

  


  In my mind, spawning ffmpeg as a child process carries much more overhead compared to calling the library functions. Am I correct ? If so, is the overhead just in RAM, or will it also affect performance ?

  


  (It's for livestreaming so the encoding is continuous)