Recherche avancée

Sur d’autres sites (4403)

• Pipe PyQt Widget Images to ffmpeg

  10 février 2024, par throwaway17434

  I have a PyQt Window with widgets that change. I want to make a video of it. I found this Answer to be very useful, however it is does not seem to be possible to use subprocess PIPE as a target in QtGui.QPixmap's save-method. I have the feeling that I should use the native QtProcess for this kind of work, but I don't know how I can PIPE the images and I can't see the errors because I can't see the standard outpur/error either. What I want to do is something like this :

  


  from PyQt4 import QtGui, QtCore
import random

app = QtGui.QApplication([])
win    = QtGui.QWidget()
layout = QtGui.QGridLayout()
win.setLayout(layout)

#picture frame
scene = QtGui.QGraphicsScene()
canvas  = QtGui.QGraphicsView(scene)
layout.addWidget(canvas,0,0)

# start button
def run():
    # set pen
    pen = QtGui.QPen(QtCore.Qt.red)
    size = canvas.size()
    
    # start seperate process
    process = QtCore.QProcess(app)
    process.start('ffmpeg',['-y', '-f', 'image2pipe', '-vcodec', 'mjpeg', '-r', '24',  '-i', '-', '-vcodec', 'mpeg4', '-qscale', '5', 'video.avi'])
    for i in range(100):
        x = random.randint(1, size.width()-1)
        y = random.randint(1, size.height()-1)        
        scene.addLine(x,y,x,y, pen=pen)
        QtGui.QPixmap.grabWidget(win).save(process, "jpeg")
        
but_run = QtGui.QPushButton("Go!")
but_run.clicked.connect(run)
layout.addWidget(but_run,1,0)

win.show()
app.exec_()


  


• Pipe PyQt Widget Images to ffmpeg

  31 octobre 2014, par throwaway17434

  I have a PyQt Window with widgets that change. I want to make a video of it. I found this Answere to be very useful, however it is does not seem to be possible to use subprocess PIPE as a target in QtGui.QPixmap’s save-method. I have the feeling that I should use the native QtProcess for this kind of work, but I don’t know how I can PIPE the images and I can’t see the errors because I can’t see the standard outpur/error either. What I want to do is something like this :

  from PyQt4 import QtGui, QtCore
  
  import random
  
  
  
  app = QtGui.QApplication([])
  
  win    = QtGui.QWidget()
  
  layout = QtGui.QGridLayout()
  
  win.setLayout(layout)
  
  
  
  #picture frame
  
  scene = QtGui.QGraphicsScene()
  
  canvas  = QtGui.QGraphicsView(scene)
  
  layout.addWidget(canvas,0,0)
  
  
  
  # start button
  
  def run():
  
      # set pen
  
      pen = QtGui.QPen(QtCore.Qt.red)
  
      size = canvas.size()
  
  
  
      # start seperate process
  
      process = QtCore.QProcess(app)
  
      process.start('ffmpeg',['-y', '-f', 'image2pipe', '-vcodec', 'mjpeg', '-r', '24',  '-i', '-', '-vcodec', 'mpeg4', '-qscale', '5', 'video.avi'])
  
      for i in range(100):
  
          x = random.randint(1, size.width()-1)
  
          y = random.randint(1, size.height()-1)        
  
          scene.addLine(x,y,x,y, pen=pen)
  
          QtGui.QPixmap.grabWidget(win).save(process, "jpeg")
  
  
  
  but_run = QtGui.QPushButton("Go!")
  
  but_run.clicked.connect(run)
  
  layout.addWidget(but_run,1,0)
  
  
  
  win.show()
  
  app.exec_()

• FFmpeg : encoding in webm (vp9) for windows

  2 juillet 2015, par Seltymar

  I’m trying to encode jpegs to a webm video in vp9 on windows 7.

  I followed the instructions from webmproject in 2 pass.
  I changed \dev\nul by NUL.

  ffmpeg.exe -i %4d.jpg -c:v libvpx-vp9 -pass 1 -b:v 1000K -threads 8 -speed 4 -tile-columns 6 -frame-parallel 1 -auto-alt-ref 1 -lag-in-frames 25 -an -f webm NUL
  
  
  
  ffmpeg.exe -i %4d.jpg  -c:v libvpx-vp9 -pass 2 -b:v 1000K -threads 8 -speed 1 -tile-columns 6 -frame-parallel 1 -auto-alt-ref 1 -lag-in-frames 25 -c:a libopus -b:a 64k -f webm out.webm

  but after the first pass, it tells me "output file is empty, nothing was encoded". But the second pass create the video.

  I also tryed in one pass :

  ffmpeg -i %4d.jpg -c:v libvpx-vp9 -crf 10 -b:v 0 output.webm

  Then I want to use the output video using html5 video and canvas tags as below.

      
  
  
  
  
  
  
  
  
  
  <canvas>
  
          Your browser does not support the canvas tag.
  
  </canvas> 
  
  <code class="echappe-js">&lt;script&gt;<br />
  
  <br />
  
  <br />
  
          var videos = document.createElement('video');<br />
  
          videos.src = &quot;output.webm&quot;;<br />
  
          videos.type = &quot;video/webm&quot;;<br />
  
          videos.load(); <br />
  
          videos.play();<br />
  
          videos.playbackRate = 1;<br />
  
          var videoImage0 = document.getElementById('canvas');        <br />
  
          videoImage0.width = 2700;<br />
  
          videoImage0.height = 1350;<br />
  
  <br />
  
          var videoImageContexts =videoImage0.getContext('2d');<br />
  
          //// background color if no video present<br />
  
          videoImageContexts.fillStyle = '#0000ff';<br />
  
          videoImageContexts.fillRect(0, 0, videoImage0.width, videoImage0.height);       <br />
  
          animate();<br />
  
  <br />
  
      function animate() {<br />
  
          requestAnimationFrame(animate);<br />
  
          update();<br />
  
  <br />
  
      }<br />
  
  <br />
  
      function update() {<br />
  
  <br />
  
          if (videos.readyState === videos.HAVE_ENOUGH_DATA) {<br />
  
              videoImageContexts.drawImage(videos, 0, 0);<br />
  
          }<br />
  
      }<br />
  
  &lt;/script&gt;
  

  result :

  1. With both methods the output video can’t be read in VLC but it can be read if I drag and drop in firefox or chrome or using the video tag only.
  2. The html page above works correctly on chrome but not in firefox. In firefox, the color is not correct like if the blue channel was translated.
     

  I think the problem is the encoding with FFmpeg. I’m using the static build from zeranoe. Here is the version

  and libvpx is enable.

  My questions :

  • Is there a problem on my command line ?

  • Is there a problem in my JS code that could lead to error in the decoding process ?