Recherche avancée

Sur d’autres sites (4930)

• JavaCV FFmpegFrameRecorder properties explanation needed

  29 décembre 2014, par Leron

  I’m using FFmpegFrameRecorder to get the video input from my webcam and record it into a video file. The problem is that I’m building my application using a few different demo source codes that I found and I use properties some of which are not completely clear to me.

  First, here is my code snippet :

  FFmpegFrameRecorder recorder = new FFmpegFrameRecorder(FILENAME,  grabber.getImageWidth(),grabber.getImageHeight());
  
  
  
          recorder.setVideoCodec(13);
  
          recorder.setFormat("mp4");
  
          recorder.setPixelFormat(avutil.PIX_FMT_YUV420P);
  
          recorder.setFrameRate(30);
  
          recorder.setVideoBitrate(10 * 1024 * 1024);
  
  
  
          recorder.start();

  • setVideoCodec(13) - What is the meaning of this (13) how can I understand what actual codec stands behind any number ?
  • setPixelFormat - Just get this, don’t know what it’s doing in general
  • setFrameRate(30) - I think this should be pretty clear but still what is the logic behind what frame rate we choose (isn’t the high the better ?)
  • setVideoBitrate(10*1024*1024) - again almost no idea what this does and what’s the logic behind the numbers ?

  At the end I just want to mention one last problem that I get recording video like this. If the actual length of the video is let’s say 20secs. When I play the video file created from the program it runs significantly faster. Can’t tell if it’s exactly 2 times faster than it should be but in general if I record a 20sec video then it’s played for about 10secs. What may cause this and how can I fix it ?

• How can i create a slideshow of some selected frames from a video in Python ?

  24 mai 2021, par s59_60r

  I have a video. In that video, there are some frames which i call Key Frames. I want to select those frames and make an separate video out of them, maybe with FPS 10, and i want to do this as efficiently as i can

  


  This is what im doing right now.

  


  I keep a list of frame numbers which i wanna extract. Then I use the following code to save the frames as image, using OpenCV

  


  myFrameNumber = frame_no
cap = cv2.VideoCapture(VIDEO_FILE)
totalFrames = cap.get(cv2.CAP_PROP_FRAME_COUNT)
if myFrameNumber >= 0 & myFrameNumber <= totalFrames:
    cap.set(1,myFrameNumber)
ret, frame = cap.read()
plt.imshow(frame)
plt.savefig(f'{RESULT_FOLDER}/{frame_no}.png')
plt.close()


  


  After this i have around 200-300 images in a result folder, now i use skvideo
to combine those images into a video, using the following code.

  


  frame = cv2.imread(os.path.join(image_folder, images[0]))
height, width, layers = frame.shape
out_video =  np.empty([len(images), height, width, 3], dtype = np.uint8)
out_video =  out_video.astype(np.uint8)
for i in range(0,len(images)):
    img = cv2.imread(image_folder+'/'+images[i])
    out_video[i] = img
skvideo.io.vwrite(video_name, out_video, inputdict={'-r':str(BASE_FPS)})


  


  but this whole process is rather slow, and i feel it can be done better. Any suggestion ? The goal is to do this thing fast and efficient

  


• FFMPEG Layering two .mov files color issue

  21 février 2023, par JDC

  I have two folders containing numbered .mov files all the same length, one folder for front and one for back. I'm using ffmpeg to combine the front and back videos with matching numbers into one output video (the front video has an alpha channel). The below command works, but in the output video the front layer has slightly incorrect colors compared to the input (reds are too yellow for example). The back layer in the output looks correct. Both input files use the bt709 color space (checked with ffprobe). Is there something in the overlay options which is causing the front layer to look incorrect, or something I'm doing wrong with the colorspace options ?

  


  


  for f in Video_Back/Video_Back_*.mov ; do
num=$(echo "$f" | sed -n 's/.*Video_Back_([0-9]5).mov$/\1/p')
if [ -f "Video_Front/Video_Front_$num.mov" ] ; then
ffmpeg -color_primaries bt709 -color_trc gamma28 -colorspace bt709 -i "$f" -color_primaries bt709 -color_trc gamma28 -colorspace bt709 -i "Video_Front/Video_Front_$num.mov" -filter_complex "[1:v]format=rgba[fg] ;[0:v][fg]overlay=shortest=1:alpha=straight,eq=gamma=1.0[mixed]" -map "[mixed]" -map 0:a:0 -c:v libx264 -crf 18 -preset veryfast -c:a copy "Video_Merged/Video_Merged_$num.mp4"
fi
done