Recherche avancée

Sur d’autres sites (11393)

• How create a circular video (transparent area on top of video) effect without applying image mask to video

  26 septembre 2024, par Arina Lubimova

  Basically i googled a lot and solutions suggests apply some PNG mask or do not provide needed solution.

  


  What i've found.

  


  ffmpeg -i main.mkv -i facecloseup.mkv
 -filter_complex "[1]trim=end_frame=1,
  geq='st(3,pow(X-(W/2),2)+pow(Y-(H/2),2));if(lte(ld(3),pow(min(W/2,H/2),2)),255,0)':128:128,
  loop=-1:1,setpts=N/FRAME_RATE/TB[mask];
  [1][mask]alphamerge[cutout];
  [0][cutout]overlay=x=W-w:y=0[v];
  [0][1]amix=2[a]"
 -map "[v]" -map "[a]"  out.mp4


  


  command = "-i " + this.video1Path.getPath() + " -i " + this.video2Path.getPath() + " -filter_complex [1]trim=end_frame=1,geq=lum_expr='st(3,pow(X-(W/2),2)+pow(Y-(H/2),2));if(lte(ld(3)," + (this.mZoomLayout.getZoomedWidth()/2) + "*" + (this.mZoomLayout.getZoomedWidth()/2) + "),255,0)':128:128,format=gray,loop=-1:1,setpts=N/FRAME_RATE/TB[mask];[1][mask]alphamerge,format=rgba,lutrgb=a=if(gte(val\\,16)\\,val)[cutout];[0][cutout]overlay=" + this.mZoomLayout.getCircleX() + ":" + this.mZoomLayout.getCircleY() + ":enable='between(t,0," + this.videoTwoDuration + ") -c:v libx264 -crf 24 -preset ultrafast " + videoPath.getPath(); 


  


  So i tried to extract needed things from them, but i don't understand how exactly i need to do that, i did this :

  


  ffmpeg -i video.mp4 -filter_complex "[0]geq='st(3,pow(X-(W/2),2)+pow(Y-(H/2),2));if(lte(ld(3),pow(min(W/2,H/2),2)),255,0)':H:W; [0:v][mask]alphamerge" out.mp4


  


  [mov,mp4,m4a,3gp,3g2,mj2 @ 000001f761dd8e40] Invalid stream specifier: mask.
    Last message repeated 1 times
Stream specifier 'mask' in filtergraph description [0]geq='st(3,pow(X-(W/2),2)+pow(Y-(H/2),2));if(lte(ld(3),pow(min(W/2,H/2),2)),255,0)':H:W; [0:v][mask]alphamerge matches no streams.


  


  ffmpeg -i video.mp4 -filter_complex "[0]geq=lum_expr='st(3,pow(X-(W/2),2)+pow(Y-(H/2),2));if(lte(ld(3),pow(min(W/2,H/2),2)),255,0)':H:W; [0:v][mask]alphamerge" out.mp4


  


  [mov,mp4,m4a,3gp,3g2,mj2 @ 000001bfd9218e80] Invalid stream specifier: mask.
    Last message repeated 1 times
Stream specifier 'mask' in filtergraph description [0]geq=lum_expr='st(3,pow(X-(W/2),2)+pow(Y-(H/2),2));if(lte(ld(3),pow(min(W/2,H/2),2)),255,0)':H:W; [0:v][mask]alphamerge matches no streams.


  


  And one more time, guys, if you are going to post some "prepared image mask" solution - just leave, the question is about creating mask on air.

  


  So, let's say we have red square (yes, ratio is static, always 1:1), yes, i can't post it because i dont have 10 rep. (...).

  


  https://i.sstatic.net/MsL71.png - red square.

  


  https://i.sstatic.net/aIFEV.png - circle

  


  https://i.sstatic.net/R8EAx.png - result

  


  https://i.sstatic.net/WtqQg.png - final result

  


  I actually want to get the answer from @Gyan or @llogan because i searched a lot and only these two guys do understand how to make things programmatically.

  


  More tech details :
Aspect ratio is constant - 1:1, width and height should be taken from the video in auto way, we need to create a white square with transparent circle inside it, the end result must contain "rounded" video with white background.

  


• Using FFMPEG, how do we add subtitles in the black bar area or under the video ?

  26 septembre 2020, par DunceDancer

  I followed these steps :

  


  


  1. Added the black bars

     


     -vf "scale=1920:1080:force_original_aspect_ratio=decrease,pad=1920:1080 :(ow-iw)/2 :(oh-ih)/2,setsar=1" Source :How to add black borders to video

     


  


  2. Added the subtitles ("burned" it into the video)

     


     ffmpeg -i "imput.mp4" -lavfi "subtitles=subtitles.srt:force_style='Alignment=0,OutlineColour=&H100000000,BorderStyle=3,Outline=1,Shadow=0,Fontsize=18,MarginL=5,MarginV=25'" -crf 1 -c:a copy "output.mp4" Source : ffmpeg subtitles alignment and position

     


  


  


  Now I am stuck as to how to place the subtitles under the video or in the black screen.

  


  Edit : Screenshot added to clarify

  


  Screenshot of the Problem

  


• dnn_backend_native_layer_mathunary : add floor support

  6 août 2020, par Mingyu Yin

  dnn_backend_native_layer_mathunary : add floor support
  It can be tested with the model generated with below python script :
  import tensorflow as tf
  
  import os
  
  import numpy as np
  
  import imageio
  
  from tensorflow.python.framework import graph_util
  
  name = 'floor'
  pb_file_path = os.getcwd()
  
  if not os.path.exists(pb_file_path+'/{}_savemodel/'.format(name)) :
  
      os.mkdir(pb_file_path+'/{}_savemodel/'.format(name))
  with tf.Session(graph=tf.Graph()) as sess :
  
      in_img = imageio.imread('detection.jpg')
  
      in_img = in_img.astype(np.float32)
  
      in_data = in_img[np.newaxis, :]
  
      input_x = tf.placeholder(tf.float32, shape=[1, None, None, 3], name='dnn_in')
  
      y_ = tf.math.floor(input_x*255)/255
  
      y = tf.identity(y_, name='dnn_out')
  
      sess.run(tf.global_variables_initializer())
  
      constant_graph = graph_util.convert_variables_to_constants(sess, sess.graph_def, ['dnn_out'])
      with tf.gfile.FastGFile(pb_file_path+'/{}_savemodel/model.pb'.format(name), mode='wb') as f :
  
          f.write(constant_graph.SerializeToString())
      print("model.pb generated, please in ffmpeg path use\n \n \
  
      python tools/python/convert.py {}_savemodel/model.pb —outdir={}_savemodel/ \n \nto generate model.model\n".format(name,name))
      output = sess.run(y, feed_dict= input_x : in_data)
  
      imageio.imsave("out.jpg", np.squeeze(output))
      print("To verify, please ffmpeg path use\n \n \
  
      ./ffmpeg -i detection.jpg -vf format=rgb24,dnn_processing=model={}_savemodel/model.pb:input=dnn_in:output=dnn_out:dnn_backend=tensorflow -f framemd5 {}_savemodel/tensorflow_out.md5\n  \
  
      or\n \
  
      ./ffmpeg -i detection.jpg -vf format=rgb24,dnn_processing=model={}_savemodel/model.pb:input=dnn_in:output=dnn_out:dnn_backend=tensorflow {}_savemodel/out_tensorflow.jpg\n \nto generate output result of tensorflow model\n".format(name, name, name, name))
      print("To verify, please ffmpeg path use\n \n \
  
      ./ffmpeg -i detection.jpg -vf format=rgb24,dnn_processing=model={}_savemodel/model.model:input=dnn_in:output=dnn_out:dnn_backend=native -f framemd5 {}_savemodel/native_out.md5\n  \
  
      or \n \
  
      ./ffmpeg -i detection.jpg -vf format=rgb24,dnn_processing=model={}_savemodel/model.model:input=dnn_in:output=dnn_out:dnn_backend=native {}_savemodel/out_native.jpg\n \nto generate output result of native model\n".format(name, name, name, name))
  Signed-off-by : Mingyu Yin <mingyu.yin@intel.com>
  

  • [DH] libavfilter/dnn/dnn_backend_native_layer_mathunary.c
  • [DH] libavfilter/dnn/dnn_backend_native_layer_mathunary.h
  • [DH] tests/dnn/dnn-layer-mathunary-test.c
  • [DH] tools/python/convert_from_tensorflow.py
  • [DH] tools/python/convert_header.py