Recherche avancée

Sur d’autres sites (9004)

• Video decoding using ffms2 (ffmpegsource)

  21 juin 2013, par praks411

  I'm using ffms2 (aka FFmpegSource) for decoding video frames and display on UI based on wxWidgets.
  My player works fine for low resolution video (320*240, 640*480) but for higher resolution (1080) it is very slow. I'm not able to meed the desired frame for high resolution video.
  After time analysis I found that FFMS_GetFrame() frame function takes much longer time for high resolution frame.
  Here are the results.
  1. 320*240 FFMS_GetFrame takes 4-6ms
  2. 640*480 FFMS_GetFrame takes >20ms
  3. 1080*720 FFMS_GetFrame takes >40

  Which means that I'll never meets 30 fps requirement for 1080p frame with FFMS2. But I'm not sure if this is the case.
  Please suggest what could be going wrong.

  void SetPosition(int64 pos)
  
  {
  
      uint8_t* data_ptr = NULL;
  
      /*check if position is valid*/
  
  
  
      if (!m_track || pos < 0  && pos > m_videoProp->NumFrames - 1)
  
          return; // ERR_POS;
  
  
  
      wxMilliClock_t start_wx_t =  wxGetLocalTimeMillis();
  
      long long start_t = start_wx_t.GetValue();
  
      m_frameId = pos;
  
     if(m_video)
  
     {
  
       m_frameProp = FFMS_GetFrame(m_video, m_frameId, &m_errInfo);
  
  
  
       if(!m_frameProp) return;
  
  
  
       if(m_frameProp)
  
       {
  
          m_width_ffms2 = m_frameProp->EncodedWidth;
  
          m_height_ffms2 = m_frameProp->EncodedHeight;
  
       }
  
  
  
         wxMilliClock_t end_wx_t =  wxGetLocalTimeMillis();
  
      long long end_t = end_wx_t.GetValue();
  
      long long diff_t =  end_t - start_t;
  
      wxLogDebug(wxString(wxT("Frame Grabe Millisec") + ToString(diff_t)));
  
  
  
      //m_frameInfo = FFMS_GetFrameInfo(m_track, FFMS_TYPE_VIDEO);
  
  
  
      /* If you want to change the output colorspace or resize the output frame size, now is the time to do it. 
  
      IMPORTANT: This step is also required to prevent resolution and colorspace changes midstream. You can 
  
      always tell a frame's original properties by examining the Encoded properties in FFMS_Frame. */
  
  
  
      /* A -1 terminated list of the acceptable output formats (see pixfmt.h for the list of pixel formats/colorspaces).
  
      To get the name of a given pixel format, strip the leading PIX_FMT_ and convert to lowercase. For example, 
  
      PIX_FMT_YUV420P becomes "yuv420p". */
  
  #if 0
  
      int pixfmt[2];
  
      pixfmt[0] = FFMS_GetPixFmt("bgr24"); 
  
      pixfmt[1] = -1;
  
  #endif
  
      // FFMS_SetOutputFormatV2 returns 0 on success. It Returns non-0 and sets ErrorMsg on failure.
  
      int failure = FFMS_SetOutputFormatV2(m_video, pixfmt, m_width_ffms2, m_height_ffms2, FFMS_RESIZER_BICUBIC, &m_errInfo);
  
      if (failure) 
  
      {
  
          //FFMS_DestroyVideoSource(m_video);
  
          //m_video = NULL;
  
          return; //return ERR_POS;
  
      }
  
       data_ptr = m_frameProp->Data[0]; 
  
  
  
    }
  
     else
  
     {
  
          m_width_ffms2 = 320;
  
          m_height_ffms2 = 240;
  
     }
  
     if(data_ptr)
  
     {
  
       memcpy(m_buf, data_ptr, 3*m_height_ffms2 * m_width_ffms2);
  
     }
  
     else
  
     {
  
       memset(m_buf, 0, 3*m_height_ffms2 * m_width_ffms2);    
  
     }
  
  }

• FFMPEG converting media type aswell as relocating MOOV atom [migrated]

  11 juin 2013, par Samuel Richards

  I have found this line online

  ffmpeg -i input.mp4 -c:a copy -c:v copy -movflags faststart output.mp4

  Which takes an input, and copies the audio, video and sets the MOOV atom to the beginning in an element called output.mp4. Basically creating the same video with the MOOV atoms in a new place.

  I need to convert the type of this media though - so I have the line

  ffmpeg -i input.wmv -vcodec h264 -f mp4 output.mp4

  My problem is, is this possible in one line ? I know I can run one after another but that creates a couple of videos when I only need the resulting video, which could result in a lot of memory wasted.

  I am sure the line

  ffmpeg -i input.wmv -movflags faststart -vcodec h264 -f mp4 output.mp4 

  Is legitimate, but without the straight copies, this takes a long long time to compute.

  Any input on MOOV atoms and ffmpeg computation is welcome.

• How to change DAR of a video to 11:9 by using FFMPEG ?

  6 novembre 2015, par Alexandr

  I want to scale video to 176x144 with DAR 11:9 but I can’t. It always sets DAR 4:3 and SAR 12:11. I use this command :

  ffmpeg -i input.mp4 -y -r 12 -filter:v "scale=iw*min(176/iw\,144/ih):ih*min(176/iw\,144/ih),pad=176:144:(176-iw*min(176/iw\,144/ih))/2:(144-ih*min(176/iw\,144/ih))/2,setdar=11:9" -vcodec h263 -b:v 55k -crf 30 -minrate:v 0k -maxrate:v 63k -bufsize:v 128k -acodec libfaac -ac 1 -ab 32k -ar 22050 output.3gp

  Please, help me understand what is going on.