Recherche avancée

Sur d’autres sites (5836)

• Video Overlay not returning to black once video stopped

  24 septembre 2020, par Matt Nelson

  I am combining multiple videos from a webrtc call and aligning the "user-terminal" videos to the left, and the "user-visitor" videos to the right. As the visitor feed can start and strop, they have multiple videos so I'm offsetting them by the timestamp.

  


  This is working, however crazy it looks !

  


  The one last issue I have is that when the first visitor video stop on the right, it shows the last frame of that video until the next video starts on the left. Can I have it return o the black background ?

  


  Here is the command passed to ffmpeg :

  


  ffmpeg -y 
-i /recordings/process/5f6c9c3/videoroom-5f6c9c3-user-terminal-1600953586531366-audio.mjr.opus 
-i /recordings/process/5f6c9c3/videoroom-5f6c9c3-user-visitor-1600953592694430-audio.mjr.opus 
-i /recordings/process/5f6c9c3/videoroom-5f6c9c3-user-visitor-1600953609873223-audio.mjr.opus 
-i /recordings/process/5f6c9c3/videoroom-5f6c9c3-user-visitor-1600953628668227-audio.mjr.opus 
-i /recordings/process/5f6c9c3/videoroom-5f6c9c3-user-visitor-1600953663905342-audio.mjr.opus 
-i /recordings/process/5f6c9c3/videoroom-5f6c9c3-user-MasterTerminal-52350116-1600953681107272-audio.mjr.opus 
-i /recordings/process/5f6c9c3/videoroom-5f6c9c3-user-visitor-1600953697832165-audio.mjr.opus 
-i /recordings/process/5f6c9c3/videoroom-5f6c9c3-user-MasterTerminal-52350116-1600953723320364-audio.mjr.opus 
-i /recordings/process/5f6c9c3/videoroom-5f6c9c3-user-visitor-1600953725307043-audio.mjr.opus 
-filter_complex [1:a]adelay=6163|6163[1adelay];[2:a]adelay=23341|23341[2adelay];[3:a]adelay=42136|42136[3adelay];[4:a]adelay=77373|77373[4adelay];[5:a]adelay=94575|94575[5adelay];[6:a]adelay=111300|111300[6adelay];[7:a]adelay=136788|136788[7adelay];[8:a]adelay=138775|138775[8adelay];[0:a][1adelay][2adelay][3adelay][4adelay][5adelay][6adelay][7adelay][8adelay]amix=inputs=9:duration=longest[a] 
-map [a] -ac -2 /recordings/process/5f6c9c3/5f6c9c3.mp3


  


• Rendering YUV420P ffmpeg decoded images on QT with OpenGL, only see black screen

  17 février 2019, par Lucas Zanella

  I’ve found this QT OpenGL Widget which should render a 420PYUV image on screen. I’m feeding a ffmpeg decoded buffer into its paintGL() function but I see nothing. Neither noises or correct images, only a black screen. I’m trying to understand why.

  I want to exclude the possibilities of other things being wrong, but I need to be sure first that my code will produce anything. I std::couted some bytes from the ffmpeg just to see if they were arriving and they were. So I should see at least some noise.

  Can you see anything wrong with my code that wouldn’t make it able to render images on screen ?

  This is the widget that should output the image :

  #include "XVideoWidget.h"
  
  #include <qdebug>
  
  #include <qtimer>
  
  #include <iostream>
  
  //自动加双引号
  
  #define GET_STR(x) #x
  
  #define A_VER 3
  
  #define T_VER 4
  
  
  
  //顶点shader
  
  const char *vString = GET_STR(
  
      attribute vec4 vertexIn;
  
      attribute vec2 textureIn;
  
      varying vec2 textureOut;
  
      void main(void)
  
      {
  
          gl_Position = vertexIn;
  
          textureOut = textureIn;
  
      }
  
  );
  
  
  
  
  
  //片元shader
  
  const char *tString = GET_STR(
  
      varying vec2 textureOut;
  
      uniform sampler2D tex_y;
  
      uniform sampler2D tex_u;
  
      uniform sampler2D tex_v;
  
      void main(void)
  
      {
  
          vec3 yuv;
  
          vec3 rgb;
  
          yuv.x = texture2D(tex_y, textureOut).r;
  
          yuv.y = texture2D(tex_u, textureOut).r - 0.5;
  
          yuv.z = texture2D(tex_v, textureOut).r - 0.5;
  
          rgb = mat3(1.0, 1.0, 1.0,
  
              0.0, -0.39465, 2.03211,
  
              1.13983, -0.58060, 0.0) * yuv;
  
          gl_FragColor = vec4(rgb, 1.0);
  
      }
  
  
  
  );
  
  
  
  
  
  
  
  //准备yuv数据
  
  // ffmpeg -i v1080.mp4 -t 10 -s 240x128 -pix_fmt yuv420p  out240x128.yuv
  
  XVideoWidget::XVideoWidget(QWidget * parent)
  
  {
  
     // setWindowFlags (Qt::WindowFullscreenButtonHint);
  
    //  showFullScreen();
  
  
  
  }
  
  
  
  XVideoWidget::~XVideoWidget()
  
  {
  
  }
  
  
  
  //初始化opengl
  
  void XVideoWidget::initializeGL()
  
  {
  
      //qDebug() &lt;&lt; "initializeGL";
  
      std::cout &lt;&lt; "initializing gl" &lt;&lt; std::endl;
  
      //初始化opengl （QOpenGLFunctions继承）函数
  
      initializeOpenGLFunctions();
  
  
  
      this->m_F  = QOpenGLContext::currentContext()->functions();
  
  
  
      //program加载shader（顶点和片元）脚本
  
      //片元（像素）
  
      std::cout &lt;&lt; program.addShaderFromSourceCode(QOpenGLShader::Fragment, tString) &lt;&lt; std::endl;
  
      //顶点shader
  
      std::cout &lt;&lt; program.addShaderFromSourceCode(QOpenGLShader::Vertex, vString) &lt;&lt; std::endl;
  
  
  
      //设置顶点坐标的变量
  
      program.bindAttributeLocation("vertexIn",A_VER);
  
  
  
      //设置材质坐标
  
      program.bindAttributeLocation("textureIn",T_VER);
  
  
  
      //编译shader
  
      std::cout &lt;&lt; "program.link() = " &lt;&lt; program.link() &lt;&lt; std::endl;
  
  
  
      std::cout &lt;&lt; "program.bind() = " &lt;&lt; program.bind() &lt;&lt; std::endl;
  
  
  
      //传递顶点和材质坐标
  
      //顶点
  
      static const GLfloat ver[] = {
  
          -1.0f,-1.0f,
  
          1.0f,-1.0f,
  
          -1.0f, 1.0f,
  
          1.0f,1.0f
  
      };
  
  
  
      //材质
  
      static const GLfloat tex[] = {
  
          0.0f, 1.0f,
  
          1.0f, 1.0f,
  
          0.0f, 0.0f,
  
          1.0f, 0.0f
  
      };
  
  
  
      //顶点
  
      glVertexAttribPointer(A_VER, 2, GL_FLOAT, 0, 0, ver);
  
      glEnableVertexAttribArray(A_VER);
  
  
  
      //材质
  
      glVertexAttribPointer(T_VER, 2, GL_FLOAT, 0, 0, tex);
  
      glEnableVertexAttribArray(T_VER);
  
  
  
      //glUseProgram(&amp;program);
  
      //从shader获取材质
  
      unis[0] = program.uniformLocation("tex_y");
  
      unis[1] = program.uniformLocation("tex_u");
  
      unis[2] = program.uniformLocation("tex_v");
  
  
  
      //创建材质
  
      glGenTextures(3, texs);
  
  
  
      //Y
  
      glBindTexture(GL_TEXTURE_2D, texs[0]);
  
      //放大过滤，线性插值   GL_NEAREST(效率高，但马赛克严重)
  
      glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
  
      glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
  
      //创建材质显卡空间
  
      glTexImage2D(GL_TEXTURE_2D, 0, GL_RED, width, height, 0, GL_RED, GL_UNSIGNED_BYTE, 0);
  
  
  
      //U
  
      glBindTexture(GL_TEXTURE_2D, texs[1]);
  
      //放大过滤，线性插值
  
      glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
  
      glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
  
      //创建材质显卡空间
  
      glTexImage2D(GL_TEXTURE_2D, 0, GL_RED, width/2, height / 2, 0, GL_RED, GL_UNSIGNED_BYTE, 0);
  
  
  
      //V
  
      glBindTexture(GL_TEXTURE_2D, texs[2]);
  
      //放大过滤，线性插值
  
      glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR);
  
      glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR);
  
      //创建材质显卡空间
  
      glTexImage2D(GL_TEXTURE_2D, 0, GL_RED, width / 2, height / 2, 0, GL_RED, GL_UNSIGNED_BYTE, 0);
  
  
  
      ///分配材质内存空间
  
      datas[0] = new unsigned char[width*height];     //Y
  
      datas[1] = new unsigned char[width*height/4];   //U
  
      datas[2] = new unsigned char[width*height/4];   //V
  
  }
  
  
  
  //刷新显示
  
  void XVideoWidget::paintGL(unsigned char**data)
  
  //void QFFmpegGLWidget::updateData(unsigned char**data)
  
  {
  
      std::cout &lt;&lt; "painting!" &lt;&lt; std::endl;
  
      memcpy(datas[0], data[0], width*height);
  
      memcpy(datas[1], data[1], width*height/4);
  
      memcpy(datas[2], data[2], width*height/4);
  
  
  
      glActiveTexture(GL_TEXTURE0);
  
      glBindTexture(GL_TEXTURE_2D, texs[0]); //0层绑定到Y材质
  
      //修改材质内容(复制内存内容)
  
      glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width, height, GL_RED, GL_UNSIGNED_BYTE, datas[0]);
  
      //与shader uni遍历关联
  
      glUniform1i(unis[0], 0);
  
  
  
  
  
      glActiveTexture(GL_TEXTURE0+1);
  
      glBindTexture(GL_TEXTURE_2D, texs[1]); //1层绑定到U材质
  
                                             //修改材质内容(复制内存内容)
  
      glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width/2, height / 2, GL_RED, GL_UNSIGNED_BYTE, datas[1]);
  
      //与shader uni遍历关联
  
      glUniform1i(unis[1],1);
  
  
  
  
  
      glActiveTexture(GL_TEXTURE0+2);
  
      glBindTexture(GL_TEXTURE_2D, texs[2]); //2层绑定到V材质
  
                                             //修改材质内容(复制内存内容)
  
      glTexSubImage2D(GL_TEXTURE_2D, 0, 0, 0, width / 2, height / 2, GL_RED, GL_UNSIGNED_BYTE, datas[2]);
  
      //与shader uni遍历关联
  
      glUniform1i(unis[2], 2);
  
  
  
      glDrawArrays(GL_TRIANGLE_STRIP,0,4);
  
      qDebug() &lt;&lt; "paintGL";
  
  }
  
  
  
  
  
  // 窗口尺寸变化
  
  void XVideoWidget::resizeGL(int width, int height)
  
  {
  
      m_F->glViewport(0, 0, width, height);
  
  
  
      qDebug() &lt;&lt; "resizeGL "&lt;code></iostream></qtimer></qdebug>

  Here’s a bit of code from my MainWindow :

  MainWindow::MainWindow(QWidget *parent):
  
      QMainWindow(parent)
  
      {
  
          FfmpegDecoder* ffmpegDecoder = new FfmpegDecoder();
  
          if(!ffmpegDecoder->Init()) {
  
              std::cout << "problem with ffmpeg decoder init"  << std::endl;
  
          } else {
  
              std::cout << "fmmpeg decoder initiated"  << std::endl;
  
          }
  
          XVideoWidget * xVideoWidget = new XVideoWidget(parent);
  
          ffmpegDecoder->setOpenGLWidget(xVideoWidget);
  
  
  
          mediaStream = new MediaStream(uri, ffmpegDecoder, videoConsumer);//= new MediaStream(uri, ffmpegDecoder, videoConsumer);
  
          //...
  
      }
  
      void MainWindow::run()
  
      {
  
          mediaStream->receiveFrame();
  
      }

  My main.cpp makes sure my window run() method runs in the background.

      MainWindow w;
  
      w.setFixedSize(1280,720);
  
      w.show();
  
      boost::thread mediaThread(&MainWindow::run, &w);
  
      std::cout << "mediaThread running"  << std::endl;

  If someone wants to view the entire code, please feel free to visit the commit I just did : https://github.com/lucaszanella/orwell/tree/bbd74e42bd42df685bacc5d51cacbee3a178689f

• convert camera-stream to a MJPEG + RTSP stream

  15 juillet 2017, par manman

  I am trying to convert a video-stream from a ubiquiti camera to a rtsp stream decoded in mjpeg. I tried it with ffserver but it didn’t work out. Straight away : A solution in windows would be more suitable in this case, if anyone knows a good windows-software to do such things, please tell me.

  Now to my setup :
  For testing purpose, i used a Ubuntu-Desktop VM and installed the ffmpeg package including ffserver via the command apt-get install ffmpeg.
  Afterwards i used the preconfigured Feed (feed1.ffm) to send the data to ffserver with ffmpeg :

  ffmpeg -i rtsp://[Camera-Url] -strict -2 http://localhost:8090/feed1.ffm

  and configured a new Stream in ffserver.conf

  <stream>
  
    Format rtsp
  
    Feed feed1.ffm
  
    VideoCodec mjpeg
  
    VideoFrameRate 5
  
    VideoIntraOnly
  
    VideoSize 352x240
  
    NoAudio
  
  </stream>

  I now tested the stream with vlc-player with following urls but none of them worked :

  rtsp://127.0.0.1/jpgvideo.sav
  
  rtsp://127.0.0.1:5454/jpgvideo.sav
  
  rtsp://127.0.0.1:5454/jpgvideo.sav.rtsp
  
  rtsp://127.0.0.1:5454/jpgvideo.rtsp

  Does someone knows why ? What I am missing here ?

  ---

  Note that another non-rtsp stream is just working fine :

  <stream>
  
    Feed feed1.ffm
  
    Format mpjpeg
  
    VideoFrameRate 5
  
    VideoIntraOnly
  
    VideoSize 352x240
  
    NoAudio
  
  </stream>

  Url :

  http://127.0.0.1:8090/test.mjpg

  ---

  In case anybody wonders :
  I am trying to get a video-stream on a SPA525G2 Cisco-IP-Phone. This is only supported to Cisco-Cameras, but according to this link it should also be possible if the stream is cisco-camera-like. (rtsp + mjpeg, 5fps)