Recherche avancée

Sur d’autres sites (8602)

• Java - RTSP save snapshot from Stream Packets

  9 août 2016, par Guerino Rodella

  I’m developing an application which requests snapshots to DVR and IP Cameras. The device I’m working on just offer RTSP requests to do so. Then I implemented the necessary RTSP methods to start receiving the stream packets and I started receiving then via UDP connection established. My doubt is, how can I save the received data to a jpeg file ? Where’s the begging and end of the image bytes received ?

  I searched a lot libraries which implement this type of service in Java, like Xuggler ( which it’s maintained no more ), javacpp-presets - has ffmpeg and opencv libraries included - I had some environment problems with it. If someone know an easy and good one which saves snapshots from the streams, let me know.

  My code :

  final long timeout = System.currentTimeMillis() + 3000;
  
  
  
  byte[] fullImage = new byte[ 1024 * 1024 ];
  
  DatagramSocket udpSocket = new DatagramSocket( 8000 );
  
  int lastByte = 0;
  
  
  
  // Skip first 2 packets because I think they are HEADERS
  
  // Since I don't know what they mean, I just print then in hexa
  
  for( int i = 0; i < 2; i++ ){
  
  
  
      byte[] buffer = new byte[ 1024 ];
  
      DatagramPacket dataPacket = new DatagramPacket( buffer, buffer.length );
  
      udpSocket.receive( dataPacket );
  
  
  
      int dataLenght = dataPacket.getLength();
  
      buffer = Arrays.copyOf( buffer, dataLenght );
  
  
  
      System.out.println( "RECEIVED[" + DatatypeConverter.printHexBinary( buffer ) + " L: " + dataLenght );
  
  
  
  }
  
  
  
  do{
  
  
  
      byte[] buffer = new byte[ 1024 ];
  
      DatagramPacket dataPacket = new DatagramPacket( fullImage, fullImage.length );
  
      udpSocket.receive( dataPacket );
  
  
  
      System.out.println( "RECEIVED: " + new String( fullImage ) );
  
  
  
      for( int i = 0; i < buffer.length; i++ ){
  
          fullImage[ i + lastByte ] = buffer[ i ];
  
          lastByte ++;
  
  
  
      }
  
  
  
  } while( System.currentTimeMillis() < timeout );
  
  // I know this timeout is wrong, I should stop after getting full image bytes

  The output :

  RECEIVED : 80E0000100004650000000006742E01FDA014016C4 L : 21
  RECEIVED : 80E00002000046500000000068CE30A480 L : 17
  RECEIVED : Tons of data from the streaming...
  RECEIVED : Tons of data from the streaming...
  RECEIVED : Tons of data from the streaming...
  [...]

  As you might suppose, the image I’m saving into a file is not readable because I’m doing it wrong. I think the header provide me some info about the next packets the server will sent me telling the start and the end of the image from the streaming. But I don’t understood them. Someone know how to solve it ? Any tips are welcome !

• Save FFMPEG mp4 output to tempfile using Paperclip

  31 mars 2021, par aminhs

  I am trying to save the mp4 output result to a tempfile when ffmpeg concatenates two mp4 input files using the Paperclip run command. After I run the command the tempfile is always empty with text/plain format.

  


  This is my code :

  


  output = Tempfile.new(["join_video_", ".mp4"], binmode: true)
Paperclip.run("ffmpeg -i /path/to/Video/one.mp4 -i /path/to/Video/two.mp4 -filter_complex \"[0:v][1:v]concat=n=2:v=1:a=0[outv]\" -map \"[outv]\" -f mp4 #{File.expand_path(output.path)}")
output.rewind
output.close


  


  I am interested in knowing how to save the mp4 output result to a tempfile from the ffmpeg concatenation rather than knowing how to concatenate two mp4 video files.

  


  Thank you in advance.

  


• Save movie about open AI gym

  30 novembre 2019, par Andres

  I make a Reinforcement learning code on python with open AI gym.After training I want save the movie about reinforcement learning but on windows this is impossible.

  There is another way to save the movie of reinforcement learning MountainCar-v0 on windows ??

  if __name__ == "__main__":
  
      environment = gym.make("MountainCar-v0")
  
      agent = QLearner(environment)
  
      learned_policy = train(agent, environment)
  
      monitor_path = "./movie_output"
  
      environment = gym.wrappers.Monitor(environment, monitor_path, force=True)

  this is the error message

  DependencyNotInstalled : Found neither the ffmpeg nor avconv executables. On OS X, you can install ffmpeg via brew install ffmpeg. On most Ubuntu variants, sudo apt-get install ffmpeg should do it. On Ubuntu 14.04, however, you’ll need to install avconv with sudo apt-get install libav-tools.