Recherche avancée

Sur d’autres sites (3607)

• FFMPEG animation with multiple images

  7 octobre 2015, par user1402732

  I am using ffmpeg command line tool for creating video by mixing images & audio files. Everything is working as expected. Problem started when I need to add some animation to the images when creating videos. i.e. adding zoom in effect to multiple images, it has filters but none of seems to be working with multiple inputs but only for single input.

  Works fine :

  ffmpeg -loop 1 -i Winter.jpg -vf "zoompan=z='min(zoom+0.0015,1.5)':d=125" \
  
  -c:v libx264  -t 5 -s "800x450" zoomin1.mp4

  But this does not :

  ffmpeg -loop 1 -framerate 1 -i bg2.jpg -framerate 1/3 -i img%03d.jpg -i audio.mp3 \
  
  -filter_complex overlay=(W-w)/2:(H-h)/2:shortest=1,format=yuv420p,zoompan=z='zoom+0.095':d=3 \
  
  -c:v libx264 -r 30 -c:a aac -strict experimental -b:a 192k -movflags +faststart \
  
  -shortest testvideo.mp4

  It’s with multiple image and background image & audio.

  It zooms but transition is not smooth and also there’s a gap between images. It should go smoothly one after another.

• How to use ffmpeg to make a video from one still image with height from 0 to 100% animation，

  30 août 2021, par Xiang Chen

  I want use a still image to make a transparent webm video, this video needs to have the following style :

  


  

  1. The image height auto increase from 0 to 100% in specific time.

  


  2. I don't want scroll effect , what i want is just like the image spread from top to bottom

  


  


  Below is my demo image :

  


  

  


  This is the effect I want :

  


  

  


  Black color part represent transparency.

  


• Python animation saves only one single frame

  5 octobre 2015, par JohnND

  I’m trying to save a python animation as a video file. The animation itself works fine with show(). But when I use save() I get a video file containing only one frame which stays on for the duration of the video. The duration itself seems correct (2 minutes).
  Here’s my script, it gets some data from files that are read by custom functions then use the data to make an animation. I’ve tried various alternatives like specifying ffmpeg as writer or using Agg as backend. So far it all gives the same result. Is there some obvious misuse of the animation module ?

  #!/usr/bin/env python
  
  
  
  import sys
  
  import os
  
  import matplotlib
  
  #matplotlib.use("Agg")
  
  import numpy as np
  
  import math as mh
  
  from pylab import *
  
  from matplotlib import animation as animation
  
  #plt.rcParams['animation.ffmpeg_path'] = '/usr/bin/ffmpeg'
  
  
  
  sys.path.append('/home/jseverin/bin/myPythonScripts/')
  
  import temp_profile_v4
  
  
  
  ##########################################################################
  
  #                          Function definition                           #
  
  ##########################################################################
  
  
  
  
  
  ##########################################################################
  
  #                                  Main                                  #
  
  ##########################################################################
  
  
  
  def main(args):
  
      log_file = args[1]
  
  
  
      lz, coords, instant_profiles, fig1, Nlayers = temp_profile_v4.main(["temp_profile_v4.py", "-i", log_file])
  
      close(fig1)
  
  
  
      # Layer thickness in angstroms
  
      thickA = lz/Nlayers
  
      thickA2 = thickA/2
  
  
  
  
  
      # We need to slice the array
  
      # We choose the width of the exclusion area as a fraction of the source/sink width
  
      fract = 1.0
  
      safeDist = fract*thickA
  
      #sliced_coords, sliced_temps = slice_prof(safeDist, thickA2, lz, slice_coords, slice_temps)
  
  
  
      Nframes = int(len(instant_profiles)/2)
  
  
  
      # Plot setup
  
  
  
      fig = figure(figsize=(12,8.5)) # initialise la figure
  
      line, = plot([],[], '-', linewidth=1.5, label = "T(z)")
  
  
  
      ymax = 0
  
      for prof in instant_profiles :
  
          dummy = max(prof)
  
          if dummy > ymax :
  
              ymax = dummy
  
  
  
      ymin = 999999999
  
      for prof in instant_profiles :
  
          dummy = min(prof)
  
          if dummy < ymin :
  
              ymin = dummy
  
  
  
      xlim([-0.1, lz])
  
      ylim([ymin, ymax])
  
      xlabel(r'z ($\AA$)', fontsize=14)
  
      ylabel('Temperature (K)', fontsize=14)
  
  
  
      # Show source and sink on graph
  
      axvspan(lz/4-thickA2, lz/4+thickA2, alpha=0.2, color='black', label=r"source/sink (a = %d $\AA$)" % int(round(thickA)))
  
      axvspan(3*lz/4-thickA2, 3*lz/4+thickA2, alpha=0.2, color='black')
  
      axvline(x=lz/4 + thickA2 + safeDist, linestyle='--', color='k')
  
      axvline(x=3*lz/4 - thickA2 - safeDist, linestyle='--', color='k')
  
  
  
      # fonction a definir quand blit=True
  
      # cree l'arriere de l'animation qui sera present sur chaque image
  
      def init():
  
          line.set_data([],[])
  
          return line,
  
  
  
      def animate(i): 
  
          step = 2*i
  
          line.set_data(coords, instant_profiles[step])
  
          return line,
  
  
  
      ani = animation.FuncAnimation(fig, animate, init_func=init, frames=Nframes, blit=True, interval=20, repeat=False)
  
  
  
      show()
  
  
  
      #FFwriter = animation.FFMpegWriter()
  
      #ani.save('dynamic_profile.mp4', writer = 'ffmpeg')
  
  
  
      # Set up formatting for the movie files
  
      #Writer = animation.writers['ffmpeg']
  
      #writer = Writer(fps=15, metadata=dict(artist='Me'), bitrate=1800)
  
      #ani.save('animation.mp4', writer=writer)
  
  
  
      ani.save('animation.mp4', fps=15)
  
  
  
  # End main
  
  
  
  if __name__=="__main__":
  
      main(sys.argv)