Recherche avancée

Sur d’autres sites (8152)

• FFMPEG crop video output is black screen, sound is good [on hold]

  11 janvier 2016, par Oum Alaa

  I cropped a video using ffmpeg command line, the output is black screen, the sound is clear and good :

  ffmpeg -i Original.mp4 -t 00:00:29 -vf "crop=634:300:0:60" Done.mp4

  ffmpeg version :

  ffmpeg version 0.8.17-4:0.8.17-0ubuntu0.12.04.1, Copyright (c) 2000-2014 the Libav developers

  built on Mar 16 2015 13:26:50 with gcc 4.6.3

  video information :

  mediainfo '--Inform=Video;%Width%x%Height%' Original.mp4
  
  634x360

  Thanks in advance

• ffmpeg : corrupt audio in usage of avfoundation

  3 janvier 2016, par Andy

  I am trying to record my desktop screen (MacBook Pro (Retina, 15-inch, Mid 2014), OSX 10.10.5) with audio. First, I have to find the devices for recording, video:1 and audio:0 in my case.

  User@User-MacBook-Pro:~/test$ ffmpeg -hide_banner -f avfoundation -list_devices true -i ""
  
  [AVFoundation input device @ 0x7ff19a422320] AVFoundation video devices:
  
  [AVFoundation input device @ 0x7ff19a422320] [0] FaceTime HD Camera
  
  [AVFoundation input device @ 0x7ff19a422320] [1] Capture screen 0
  
  [AVFoundation input device @ 0x7ff4b9c22320] [2] Capture screen 1
  
  [AVFoundation input device @ 0x7ff19a422320] AVFoundation audio devices:
  
  [AVFoundation input device @ 0x7ff19a422320] [0] Soundflower (2ch)
  
  [AVFoundation input device @ 0x7ff19a422320] [1] Built-in Microphone
  
  [AVFoundation input device @ 0x7ff19a422320] [2] Soundflower (64ch)
  
  : Input/output error

  Then I run to capture the screen :

  ffmpeg -f avfoundation -i "1:0" out_desktop.mkv

  and the audio of the output stutters terribly.
  But if I use the second external screen (which is connect via HDMI) :

  ffmpeg -f avfoundation -i "2:0" out_ext.mkv

  Then I have no audio problems.

  So, what could be the cause of this behaviour ? Somehow the video device "Capture screen 0" causes the audio stutter.

  Here are the full outputs of the two commands above, Screen 0 :

  User@User-MacBook-Pro:~/test$ ffmpeg -f avfoundation -i "1:0" out.mkv
  
  ffmpeg version 2.8.4 Copyright (c) 2000-2015 the FFmpeg developers
  
    built with Apple LLVM version 7.0.2 (clang-700.1.81)
  
    configuration: --prefix=/usr/local/Cellar/ffmpeg/2.8.4 --enable-shared --enable-pthreads --enable-gpl --enable-version3 --enable-hardcoded-tables --enable-avresample --cc=clang --host-cflags= --host-ldflags= --enable-opencl --enable-libx264 --enable-libmp3lame --enable-libvo-aacenc --enable-libxvid --enable-libfreetype --enable-libtheora --enable-libvorbis --enable-libvpx --enable-librtmp --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libass --enable-ffplay --enable-libspeex --enable-libschroedinger --enable-libfdk-aac --enable-libopus --enable-frei0r --enable-libopenjpeg --disable-decoder=jpeg2000 --extra-cflags=-I/usr/local/Cellar/openjpeg/1.5.2_1/include/openjpeg-1.5 --enable-nonfree --enable-vda
  
    libavutil      54. 31.100 / 54. 31.100
  
    libavcodec     56. 60.100 / 56. 60.100
  
    libavformat    56. 40.101 / 56. 40.101
  
    libavdevice    56.  4.100 / 56.  4.100
  
    libavfilter     5. 40.101 /  5. 40.101
  
    libavresample   2.  1.  0 /  2.  1.  0
  
    libswscale      3.  1.101 /  3.  1.101
  
    libswresample   1.  2.101 /  1.  2.101
  
    libpostproc    53.  3.100 / 53.  3.100
  
  [avfoundation @ 0x7fca1200da00] Selected pixel format (yuv420p) is not supported by the input device.
  
  [avfoundation @ 0x7fca1200da00] Supported pixel formats:
  
  [avfoundation @ 0x7fca1200da00]   uyvy422
  
  [avfoundation @ 0x7fca1200da00]   yuyv422
  
  [avfoundation @ 0x7fca1200da00]   nv12
  
  [avfoundation @ 0x7fca1200da00]   0rgb
  
  [avfoundation @ 0x7fca1200da00]   bgr0
  
  [avfoundation @ 0x7fca1200da00] Overriding selected pixel format to use uyvy422 instead.
  
  [avfoundation @ 0x7fca1200da00] Stream #0: not enough frames to estimate rate; consider increasing probesize
  
  Input #0, avfoundation, from '1:0':
  
    Duration: N/A, start: 11734.092500, bitrate: 2822 kb/s
  
      Stream #0:0: Video: rawvideo (UYVY / 0x59565955), uyvy422, 2880x1800, 1000k tbr, 1000k tbn, 1000k tbc
  
      Stream #0:1: Audio: pcm_f32le, 44100 Hz, stereo, flt, 2822 kb/s
  
  File 'out.mkv' already exists. Overwrite ? [y/N] y
  
  No pixel format specified, yuv422p for H.264 encoding chosen.
  
  Use -pix_fmt yuv420p for compatibility with outdated media players.
  
  [libx264 @ 0x7fca1390e000] MB rate (20340000000) > level limit (2073600)
  
  [libx264 @ 0x7fca1390e000] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX FMA3 AVX2 LZCNT BMI2
  
  [libx264 @ 0x7fca1390e000] profile High 4:2:2, level 5.2, 4:2:2 8-bit
  
  [libx264 @ 0x7fca1390e000] 264 - core 148 r2601 a0cd7d3 - H.264/MPEG-4 AVC codec - Copyleft 2003-2015 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=12 lookahead_threads=2 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=25 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
  
  Output #0, matroska, to 'out.mkv':
  
    Metadata:
  
      encoder         : Lavf56.40.101
  
      Stream #0:0: Video: h264 (libx264) (H264 / 0x34363248), yuv422p, 2880x1800, q=-1--1, 1000k fps, 1k tbn, 1000k tbc
  
      Metadata:
  
        encoder         : Lavc56.60.100 libx264
  
      Stream #0:1: Audio: vorbis (libvorbis) (oV[0][0] / 0x566F), 44100 Hz, stereo, fltp
  
      Metadata:
  
        encoder         : Lavc56.60.100 libvorbis
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (rawvideo (native) -> h264 (libx264))
  
    Stream #0:1 -> #0:1 (pcm_f32le (native) -> vorbis (libvorbis))
  
  Press [q] to stop, [?] for help
  
  frame=  969 fps= 29 q=-1.0 Lsize=    6911kB time=00:00:34.80 bitrate=1626.8kbits/s    
  
  video:6475kB audio:396kB subtitle:0kB other streams:0kB global headers:4kB muxing overhead: 0.591309%
  
  [libx264 @ 0x7fca1390e000] frame I:4     Avg QP:15.77  size:1461092
  
  [libx264 @ 0x7fca1390e000] frame P:246   Avg QP:16.00  size:  2608
  
  [libx264 @ 0x7fca1390e000] frame B:719   Avg QP:13.91  size:   200
  
  [libx264 @ 0x7fca1390e000] consecutive B-frames:  0.9%  0.4%  0.0% 98.7%
  
  [libx264 @ 0x7fca1390e000] mb I  I16..4: 11.6% 43.3% 45.2%
  
  [libx264 @ 0x7fca1390e000] mb P  I16..4:  0.0%  0.0%  0.1%  P16..4:  0.4%  0.0%  0.0%  0.0%  0.0%    skip:99.5%
  
  [libx264 @ 0x7fca1390e000] mb B  I16..4:  0.0%  0.0%  0.0%  B16..8:  0.2%  0.0%  0.0%  direct: 0.0%  skip:99.8%  L0:39.9% L1:60.0% BI: 0.1%
  
  [libx264 @ 0x7fca1390e000] 8x8 transform intra:41.6% inter:38.8%
  
  [libx264 @ 0x7fca1390e000] coded y,uvDC,uvAC intra: 76.9% 77.9% 75.5% inter: 0.0% 0.1% 0.0%
  
  [libx264 @ 0x7fca1390e000] i16 v,h,dc,p: 51% 34% 11%  4%
  
  [libx264 @ 0x7fca1390e000] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 22% 17% 19%  7%  4%  3%  8%  5% 16%
  
  [libx264 @ 0x7fca1390e000] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 21% 24% 10%  7%  6%  5%  8%  7% 12%
  
  [libx264 @ 0x7fca1390e000] i8c dc,h,v,p: 59% 14% 18%  9%
  
  [libx264 @ 0x7fca1390e000] Weighted P-Frames: Y:0.0% UV:0.0%
  
  [libx264 @ 0x7fca1390e000] ref P L0: 71.5%  3.4% 13.0% 12.1%
  
  [libx264 @ 0x7fca1390e000] ref B L0: 46.6% 52.3%  1.1%
  
  [libx264 @ 0x7fca1390e000] ref B L1: 95.1%  4.9%
  
  [libx264 @ 0x7fca1390e000] kb/s:1525.48

  and Screen 1 :

  User@User-MacBook-Pro:~/test$ ffmpeg -f avfoundation -i "2:0" out.mkv
  
  ffmpeg version 2.8.4 Copyright (c) 2000-2015 the FFmpeg developers
  
    built with Apple LLVM version 7.0.2 (clang-700.1.81)
  
    configuration: --prefix=/usr/local/Cellar/ffmpeg/2.8.4 --enable-shared --enable-pthreads --enable-gpl --enable-version3 --enable-hardcoded-tables --enable-avresample --cc=clang --host-cflags= --host-ldflags= --enable-opencl --enable-libx264 --enable-libmp3lame --enable-libvo-aacenc --enable-libxvid --enable-libfreetype --enable-libtheora --enable-libvorbis --enable-libvpx --enable-librtmp --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libass --enable-ffplay --enable-libspeex --enable-libschroedinger --enable-libfdk-aac --enable-libopus --enable-frei0r --enable-libopenjpeg --disable-decoder=jpeg2000 --extra-cflags=-I/usr/local/Cellar/openjpeg/1.5.2_1/include/openjpeg-1.5 --enable-nonfree --enable-vda
  
    libavutil      54. 31.100 / 54. 31.100
  
    libavcodec     56. 60.100 / 56. 60.100
  
    libavformat    56. 40.101 / 56. 40.101
  
    libavdevice    56.  4.100 / 56.  4.100
  
    libavfilter     5. 40.101 /  5. 40.101
  
    libavresample   2.  1.  0 /  2.  1.  0
  
    libswscale      3.  1.101 /  3.  1.101
  
    libswresample   1.  2.101 /  1.  2.101
  
    libpostproc    53.  3.100 / 53.  3.100
  
  [avfoundation @ 0x7fc2e900da00] Selected pixel format (yuv420p) is not supported by the input device.
  
  [avfoundation @ 0x7fc2e900da00] Supported pixel formats:
  
  [avfoundation @ 0x7fc2e900da00]   uyvy422
  
  [avfoundation @ 0x7fc2e900da00]   yuyv422
  
  [avfoundation @ 0x7fc2e900da00]   nv12
  
  [avfoundation @ 0x7fc2e900da00]   0rgb
  
  [avfoundation @ 0x7fc2e900da00]   bgr0
  
  [avfoundation @ 0x7fc2e900da00] Overriding selected pixel format to use uyvy422 instead.
  
  [avfoundation @ 0x7fc2e900da00] Stream #0: not enough frames to estimate rate; consider increasing probesize
  
  Input #0, avfoundation, from '2:0':
  
    Duration: N/A, start: 11809.126333, bitrate: 2822 kb/s
  
      Stream #0:0: Video: rawvideo (UYVY / 0x59565955), uyvy422, 2560x1080, 1000k tbr, 1000k tbn, 1000k tbc
  
      Stream #0:1: Audio: pcm_f32le, 44100 Hz, stereo, flt, 2822 kb/s
  
  File 'out.mkv' already exists. Overwrite ? [y/N] y
  
  No pixel format specified, yuv422p for H.264 encoding chosen.
  
  Use -pix_fmt yuv420p for compatibility with outdated media players.
  
  [libx264 @ 0x7fc2ea82ac00] MB rate (10880000000) > level limit (2073600)
  
  [libx264 @ 0x7fc2ea82ac00] using cpu capabilities: MMX2 SSE2Fast SSSE3 SSE4.2 AVX FMA3 AVX2 LZCNT BMI2
  
  [libx264 @ 0x7fc2ea82ac00] profile High 4:2:2, level 5.2, 4:2:2 8-bit
  
  [libx264 @ 0x7fc2ea82ac00] 264 - core 148 r2601 a0cd7d3 - H.264/MPEG-4 AVC codec - Copyleft 2003-2015 - http://www.videolan.org/x264.html - options: cabac=1 ref=3 deblock=1:0:0 analyse=0x3:0x113 me=hex subme=7 psy=1 psy_rd=1.00:0.00 mixed_ref=1 me_range=16 chroma_me=1 trellis=1 8x8dct=1 cqm=0 deadzone=21,11 fast_pskip=1 chroma_qp_offset=-2 threads=12 lookahead_threads=2 sliced_threads=0 nr=0 decimate=1 interlaced=0 bluray_compat=0 constrained_intra=0 bframes=3 b_pyramid=2 b_adapt=1 b_bias=0 direct=1 weightb=1 open_gop=0 weightp=2 keyint=250 keyint_min=25 scenecut=40 intra_refresh=0 rc_lookahead=40 rc=crf mbtree=1 crf=23.0 qcomp=0.60 qpmin=0 qpmax=69 qpstep=4 ip_ratio=1.40 aq=1:1.00
  
  Output #0, matroska, to 'out.mkv':
  
    Metadata:
  
      encoder         : Lavf56.40.101
  
      Stream #0:0: Video: h264 (libx264) (H264 / 0x34363248), yuv422p, 2560x1080, q=-1--1, 1000k fps, 1k tbn, 1000k tbc
  
      Metadata:
  
        encoder         : Lavc56.60.100 libx264
  
      Stream #0:1: Audio: vorbis (libvorbis) (oV[0][0] / 0x566F), 44100 Hz, stereo, fltp
  
      Metadata:
  
        encoder         : Lavc56.60.100 libvorbis
  
  Stream mapping:
  
    Stream #0:0 -> #0:0 (rawvideo (native) -> h264 (libx264))
  
    Stream #0:1 -> #0:1 (pcm_f32le (native) -> vorbis (libvorbis))
  
  Press [q] to stop, [?] for help
  
  frame=  755 fps= 30 q=-1.0 Lsize=    1025kB time=00:00:26.65 bitrate= 315.1kbits/s    
  
  video:680kB audio:326kB subtitle:0kB other streams:0kB global headers:4kB muxing overhead: 2.003084%
  
  [libx264 @ 0x7fc2ea82ac00] frame I:4     Avg QP:13.22  size:145661
  
  [libx264 @ 0x7fc2ea82ac00] frame P:191   Avg QP:19.87  size:   250
  
  [libx264 @ 0x7fc2ea82ac00] frame B:560   Avg QP:28.31  size:   116
  
  [libx264 @ 0x7fc2ea82ac00] consecutive B-frames:  1.1%  0.0%  0.4% 98.5%
  
  [libx264 @ 0x7fc2ea82ac00] mb I  I16..4: 54.5% 29.2% 16.2%
  
  [libx264 @ 0x7fc2ea82ac00] mb P  I16..4:  0.2%  0.1%  0.0%  P16..4:  0.1%  0.0%  0.0%  0.0%  0.0%    skip:99.6%
  
  [libx264 @ 0x7fc2ea82ac00] mb B  I16..4:  0.0%  0.0%  0.0%  B16..8:  0.3%  0.0%  0.0%  direct: 0.0%  skip:99.7%  L0:69.2% L1:30.8% BI: 0.0%
  
  [libx264 @ 0x7fc2ea82ac00] 8x8 transform intra:31.7% inter:13.1%
  
  [libx264 @ 0x7fc2ea82ac00] coded y,uvDC,uvAC intra: 10.9% 8.8% 8.5% inter: 0.0% 0.0% 0.0%
  
  [libx264 @ 0x7fc2ea82ac00] i16 v,h,dc,p: 69% 31%  0%  0%
  
  [libx264 @ 0x7fc2ea82ac00] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 27%  5% 67%  0%  0%  0%  0%  0%  0%
  
  [libx264 @ 0x7fc2ea82ac00] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 38% 25% 14%  3%  3%  4%  5%  4%  5%
  
  [libx264 @ 0x7fc2ea82ac00] i8c dc,h,v,p: 92%  3%  4%  0%
  
  [libx264 @ 0x7fc2ea82ac00] Weighted P-Frames: Y:0.0% UV:0.0%
  
  [libx264 @ 0x7fc2ea82ac00] ref P L0: 76.3%  1.7% 16.8%  5.2%
  
  [libx264 @ 0x7fc2ea82ac00] ref B L0: 38.9% 56.9%  4.2%
  
  [libx264 @ 0x7fc2ea82ac00] ref B L1: 97.3%  2.7%
  
  [libx264 @ 0x7fc2ea82ac00] kb/s:208.84

  EDIT

  The main differences between the two (lengthy) outputs above via

  wdiff -n output1 output2 |colordiff

  is the resolution ([-2880x1800,-] {+2560x1080,+}) and the resulting MB rate ([-(20340000000)-] {+(10880000000)+}). So, what could increase the performance of ffmpeg ?

• ffmpeg Bmp to yuv : Crash at sws_scale

  28 décembre 2015, par the-owner

  The context :
  I have a succession of continuous bitmap and I want to encode them into a light video format.
  I use ffmpeg version 2.8.3 (the build here), under qt5, qt IDE, and msvc2013 for win32.

  The problem :
  My code crash at sws_scale () (and sometimes at avcodec_encode_video2()). When I explore the stack, the crash event occurs at sws_getCachedContext (). (I can only see the stack with these ffmpeg builds).
  I only use these ffmpeg libraries (from the Qt .pro file) :

  LIBS += -lavcodec -lavformat -lswscale -lavutil

  It’s swscale which bug. And this is the code :

  void newVideo ()
  
  {
  
      ULONG_PTR gdiplusToken;
  
      GdiplusStartupInput gdiplusStartupInput;
  
      GdiplusStartup(&gdiplusToken, &gdiplusStartupInput, NULL);
  
  
  
      initBitmap (); //init bmp
  
      int screenWidth =  bmp.bmiHeader.biWidth;
  
      int screenHeight = bmp.bmiHeader.biHeight;
  
  
  
      AVCodec * codec;
  
      AVCodecContext * c = NULL;
  
      uint8_t * outbuf;
  
      int i, out_size, outbuf_size;
  
  
  
  
  
      avcodec_register_all();
  
  
  
      qDebug () << "Video encoding\n";
  
  
  
      // Find the mpeg1 video encoder
  
      codec = avcodec_find_encoder(AV_CODEC_ID_H264);
  
      if (!codec)
  
      {
  
          qDebug () << "Codec not found\n";
  
          avcodec_close(c);
  
          av_free(c);
  
          return;
  
      }
  
      else
  
          qDebug () << "H264 codec found\n";
  
  
  
      c = avcodec_alloc_context3(codec);
  
  
  
      c->bit_rate = 1000000;
  
      c->width = 800; // resolution must be a multiple of two (1280x720),(1900x1080),(720x480)
  
      c->height = 600;
  
      c->time_base.num = 1; // framerate numerator
  
      c->time_base.den = 25; // framerate denominator
  
      c->gop_size = 30; // emit one intra frame every ten frames
  
      c->max_b_frames = 1; // maximum number of b-frames between non b-frames
  
      c->pix_fmt = AV_PIX_FMT_YUV420P; //Converstion RGB to YUV ?
  
      c->codec_id = AV_CODEC_ID_H264;
  
  
  
      struct SwsContext* fooContext = sws_getContext(screenWidth, screenHeight,
  
                                                     AV_PIX_FMT_RGB32,
  
                                                     c->width, c->height,
  
                                                     AV_PIX_FMT_YUV420P,
  
                                                     SWS_FAST_BILINEAR,
  
                                                     NULL, NULL, NULL);
  
  
  
      // Open the encoder
  
      if (avcodec_open2(c, codec, NULL) < 0)
  
      {
  
          qDebug () << "Could not open codec\n";
  
          avcodec_close(c);
  
          av_free(c);
  
          return;
  
      }
  
      else qDebug () << "H264 codec opened\n";
  
  
  
      outbuf_size = 100000 + c->width*c->height*(32>>3);//*(32>>3); // alloc image and output buffer
  
      outbuf = static_cast(malloc(outbuf_size));
  
      qDebug() << "Setting buffer size to: " << outbuf_size << "\n";
  
  
  
      FILE* f = fopen("TEST.mpg","wb");
  
      if(!f) qDebug() << "x - Cannot open video file for writing\n";
  
      else qDebug() << "Opened video file for writing\n";
  
  
  
      // encode 5 seconds of video
  
      for (i = 0; i < STREAM_FRAME_RATE*STREAM_DURATION; i++) //the stop condition i < 5.0*5
  
      {
  
          qDebug () << "i = " << i;
  
          fflush(stdout);
  
  
  
          HBITMAP hBmp;
  
          if (GetScreen(hBmp) == -1) return;
  
          BYTE * pPixels;// = new BYTE [bmp.bmiHeader.biSizeImage];
  
          pPixels = getPixels (hBmp);
  
          DeleteObject (hBmp);
  
  
  
          int nbytes = avpicture_get_size(AV_PIX_FMT_YUV420P, c->width, c->height);
  
          uint8_t* outbuffer = (uint8_t*)av_malloc(nbytes*sizeof(uint8_t));
  
          if(!outbuffer) // check if(outbuf) instead
  
          {
  
              qDebug () << "Bytes cannot be allocated";
  
              return;
  
          }
  
  
  
          AVFrame* inpic = avcodec_alloc_frame(); //av_frame_alloc () ?
  
          AVFrame* outpic = avcodec_alloc_frame();
  
  
  
          outpic->pts = (int64_t)((float)i * (1000.0/((float)(c->time_base.den))) * 90);
  
          if (avpicture_fill((AVPicture*) inpic, (uint8_t*) pPixels, AV_PIX_FMT_RGB32,
  
                         screenWidth, screenHeight) < 0)
  
              qDebug () <<  "avpicture_fill Fill picture with image failed"; //Fill picture with image
  
  
  
          if(avpicture_fill((AVPicture*) outpic, outbuffer, AV_PIX_FMT_YUV420P,
  
                         c->width, c->height) < 0)
  
              qDebug () <<  "avpicture_fill failed";
  
  
  
          if (av_image_alloc(outpic->data, outpic->linesize, c->width, c->height,
  
                         c->pix_fmt, 1) < 0)
  
              qDebug () <<  "av_image_alloc failed";
  
  
  
          inpic->data[0] += inpic->linesize[0]*(screenHeight - 1); // Flipping frame
  
          inpic->linesize[0] = -inpic->linesize[0]; // Flipping frame
  
  
  
  ////////////////////////////HERE THE BUG////////////////////////////////
  
          sws_scale(fooContext,
  
                    inpic->data, inpic->linesize,
  
                    0, c->height,
  
                    outpic->data, outpic->linesize); //HERE THE BUG
  
  
  
          av_free_packet((AVPacket *)outbuf);
  
          // encode the image
  
          out_size = avcodec_encode_video2 (c, (AVPacket *) outbuf,
  
                                            (AVFrame *) outbuf_size, (int *) outpic);
  
  ///////////////////////THE CODE DONT GO BEYOND/////////////////////////////////
  
  
  
          qDebug () << "Encoding frame" << i <<" (size=" << out_size <<"\n";
  
          fwrite(outbuf, 1, out_size, f);
  
          delete [] pPixels;
  
          av_free(outbuffer);
  
          av_free(inpic);
  
          av_freep(outpic);
  
      }
  
  
  
      // get the delayed frames
  
      for(; out_size; i++)
  
      {
  
          fflush(stdout);
  
          out_size = avcodec_encode_video2 (c, (AVPacket *) outbuf,
  
                                            (AVFrame *) outbuf_size, NULL);
  
          qDebug () << "Writing frame" << i <<" (size=" << out_size <<"\n";
  
          fwrite(outbuf, 1, out_size, f);
  
      }
  
  
  
      // add sequence end code to have a real mpeg file
  
      outbuf[0] = 0x00;
  
      outbuf[1] = 0x00;
  
      outbuf[2] = 0x01;
  
      outbuf[3] = 0xb7;
  
      fwrite(outbuf, 1, 4, f);
  
      fclose(f);
  
  
  
      avcodec_close(c);
  
      free(outbuf);
  
      av_free(c);
  
      qDebug () << "Closed codec and Freed\n";
  
  }

  And the output :

  Video encoding
  
  
  
  H264 codec found
  
  
  
  H264 codec opened
  
  
  
  Setting buffer size to:  2020000 
  
  
  
  Opened video file for writing
  
  
  
  i =  0
  
  **CRASH**

  I have thougth that my bitmap wasn’t good so I have crafted a bitmap just for testing, the code was :

      uint8_t* pPixels = new uint8_t[Width * 3 * Height];
  
      int x = 50;
  
      for(unsigned int i = 0; i < Width * 3 * Height; i = i + 3) // loop for generating color changing images
  
      {
  
          pPixels [i] = x % 255; //R
  
          pPixels [i + 1] = (x) % 255; //G
  
          pPixels [i + 2] = (255 - x) % 255; //B
  
      }

  However the crash continue. Perhaps, it might prove that it’s not the bitmap (pPixels) which has a problem.

  If anyone know, why I get this bug : Maybe don’t I set one parameter well ? Or one ffmpeg deprecated function ? etc.

  ---

  EDIT 1 27/12/15

  Thanks to Ronald S. Bultje The function sws_scale () does not crash with this code, however I get an error from it bad dst image pointers. My code :

  //DESTINATION FRAME            
  
  if (avpicture_alloc ((AVPicture*) dst_frame, AV_PIX_FMT_YUV420P, c->width, c->height) < 0)
  
              {
  
                  qDebug () <<  "# avpicture_alloc failed";
  
                  return;
  
              }
  
              if(avpicture_fill((AVPicture*) dst_frame, NULL, AV_PIX_FMT_YUV420P,
  
                             c->width, c->height) < 0)
  
                  qDebug () <<  "avpicture_fill failed";
  
              avcodec_align_dimensions2 (c, &c->width, &c->height, dst_frame->linesize);
  
  
  
  //SOURCE FRAME
  
              if (avpicture_fill((AVPicture*) src_frame, (uint8_t *) pPixels, AV_PIX_FMT_RGB32,
  
                                 tmp_screenWidth, tmp_screenHeight) < 0)
  
                  qDebug () <<  "# avpicture_fill Fill picture with image failed"; //Fill picture with image
  
              avcodec_align_dimensions2 (c, &tmp_screenWidth, &tmp_screenHeight, src_frame->linesize);
  
  
  
              struct SwsContext* conversionContext = sws_getContext(tmp_screenWidth,tmp_screenHeight,AV_PIX_FMT_RGB32,c->width, c->height, AV_PIX_FMT_YUV420P,SWS_FAST_BILINEAR, NULL, NULL, NULL);
  
  
  
              int output_Height = sws_scale(conversionContext,
  
                                            src_frame->data, src_frame->linesize,
  
                                            0, tmp_screenHeight,
  
                                            dst_frame->data, dst_frame->linesize); //return 0 -> bad dst image pointers error

  ---

  EDIT 2 28/12/15

  I have tried to follow the Ronald S. Bultje’s suggestion and now I get a bad src image pointers error, I have investigated and worked many hours but I do not find a solution. Here, there is the new snippet :

  AVFrame* src_frame = av_frame_alloc ();
  
  AVFrame* dst_frame = av_frame_alloc ();
  
  AVFrame* tmp_src_frame = av_frame_alloc ();
  
  
  
  /*........I do not use them until this snippet..........*/
  
  //DESTINATION
  
  //avpicture_free ((AVPicture*)dst_frame);
  
  avcodec_align_dimensions2 (c, &c->width, &c->height, dst_frame->linesize);
  
  if (avpicture_alloc ((AVPicture*) dst_frame, AV_PIX_FMT_YUV420P, c->width, c->height) < 0)
  
  {
  
      qDebug () <<  "# avpicture_alloc failed";
  
      return;
  
  }
  
  
  
  //SOURCE
  
  //stride = src_frame->linesize [0] = ((((screenWidth * bitPerPixel) + 31) & ~31) >> 3); do I need to do that ?
  
  //== stride - I have gotten this formula from : https://msdn.microsoft.com/en-us/library/windows/desktop/dd318229(v=vs.85).aspx
  
  if (avpicture_fill((AVPicture*) src_frame, (uint8_t *) pPixels, AV_PIX_FMT_RGB32,
  
                     screenWidth, screenHeight) < 0)
  
      qDebug () <<  "# avpicture_fill Fill picture with image failed"; //Fill picture with image
  
  //linesize [0] == 21760 like commented stride
  
  
  
  //Source TO TMP Source
  
  avcodec_align_dimensions2 (c, &tmp_screenWidth, &tmp_screenHeight, tmp_src_frame->linesize);
  
  if (avpicture_fill((AVPicture*) tmp_src_frame, NULL, AV_PIX_FMT_RGB32,
  
                     tmp_screenWidth, tmp_screenHeight) < 0)
  
      qDebug () <<  "# avpicture_fill Fill picture with image failed"; //Fill picture with image
  
  
  
  av_picture_copy ((AVPicture*) tmp_src_frame, (const AVPicture*) src_frame, AV_PIX_FMT_RGB32,
  
                   screenWidth, screenHeight);
  
  
  
  struct SwsContext* conversionContext = sws_getContext(tmp_screenWidth, tmp_screenHeight,
  
                                                        AV_PIX_FMT_RGB32,
  
                                                        c->width, c->height,
  
                                                        AV_PIX_FMT_YUV420P,
  
                                                        SWS_FAST_BILINEAR,
  
                                                        NULL, NULL, NULL);
  
  
  
  int output_Height = sws_scale(conversionContext,
  
                                tmp_src_frame->data, tmp_src_frame->linesize,
  
                                0, tmp_screenHeight,
  
                                dst_frame->data, dst_frame->linesize);
  
  //ffmpeg error = bad src image pointers
  
  // output_Height == 0

  EDIT 3

  For temp Picture I have done an avcode_align_dimension2() then a avpicture_alloc() for allocating memory and avpicture_fill() in order to fill the picture pointer. Below the updated code :

  //DESTINATION
  
  //avpicture_free ((AVPicture*)dst_frame);
  
  avcodec_align_dimensions2 (c, &c->width, &c->height, dst_frame->linesize);
  
  if (avpicture_alloc ((AVPicture*) dst_frame, AV_PIX_FMT_YUV420P, c->width, c->height) < 0)
  
  {
  
      qDebug () <<  "# avpicture_alloc failed";
  
      return;
  
  }
  
  
  
  //SOURCE
  
  //src_frame->linesize [0] = ((((screenWidth * bpp) + 31) & ~31) >> 3);
  
  //src_frame->linesize [0] = stride;
  
  if (avpicture_fill((AVPicture*) src_frame, (uint8_t *) pPixels, AV_PIX_FMT_RGB32,
  
                     screenWidth, screenHeight) < 0)
  
      qDebug () <<  "# avpicture_fill Fill picture with image failed"; //Fill picture with image
  
  
  
  //Source TO TMP Source
  
  avcodec_align_dimensions2 (c, &tmp_screenWidth, &tmp_screenHeight, tmp_src_frame->linesize);
  
  if (avpicture_alloc ((AVPicture*) tmp_src_frame, AV_PIX_FMT_RGB32, tmp_screenWidth, tmp_screenHeight) < 0)
  
  {
  
      qDebug () <<  "# avpicture_alloc failed";
  
      return;
  
  }
  
  int outbuf_size = tmp_screenWidth*tmp_screenHeight*4;// alloc image and output buffer
  
  outbuf = static_cast(malloc(outbuf_size));
  
  if (avpicture_fill((AVPicture*) tmp_src_frame, outbuf, AV_PIX_FMT_RGB32,
  
                     tmp_screenWidth, tmp_screenHeight) < 0)
  
      qDebug () <<  "# avpicture_fill Fill picture with image failed"; //Fill picture with image
  
  av_picture_copy ((AVPicture*) tmp_src_frame, (const AVPicture*) src_frame, AV_PIX_FMT_RGB32,
  
                   tmp_screenWidth, tmp_screenHeight);
  
  
  
  struct SwsContext* conversionContext = sws_getContext(tmp_screenWidth, tmp_screenHeight,
  
                                                        AV_PIX_FMT_RGB32,
  
                                                        c->width, c->height,
  
                                                        AV_PIX_FMT_YUV420P,
  
                                                        SWS_FAST_BILINEAR,
  
                                                        NULL, NULL, NULL);
  
  
  
  int output_Height = sws_scale(conversionContext,
  
                                tmp_src_frame->data, tmp_src_frame->linesize,
  
                                0, tmp_screenHeight,
  
                                dst_frame->data, dst_frame->linesize);

  The call stack is as follow : av_picture_copy() is called then av_image_copy() then _VEC_memcpy() then fastcopy_I() and crash ... The problem is not the dimensions (tmp_screenWidth/Height) ? (With av_picture_copy () could we copy a picture P1 with dim W1xH1 to a picture P2 with dimension W2xH2 ?)

  EDIT 4

  Crash at av_picture_copy() which call _aligned_malloc() then av_image_copy _VEC_memcpy() and fastcopy_I()

  //SOURCE
  
  if (avpicture_fill((AVPicture*) src_frame, (uint8_t *) pPixels, AV_PIX_FMT_RGB32,
  
                     screenWidth, screenHeight) < 0)
  
      qDebug () <<  "# avpicture_fill Fill picture with image failed"; //Fill picture with image
  
  
  
  //Source TO TMP Source
  
  avcodec_align_dimensions2 (c, &tmp_screenWidth, &tmp_screenHeight, tmp_src_frame->linesize);
  
  if (avpicture_alloc ((AVPicture*) tmp_src_frame, AV_PIX_FMT_RGB32, tmp_screenWidth, tmp_screenHeight) < 0)
  
  {
  
      qDebug () <<  "# avpicture_alloc failed";
  
      return;
  
  }
  
  av_picture_copy ((AVPicture*) tmp_src_frame, (const AVPicture*) src_frame, AV_PIX_FMT_RGB32,
  
                   tmp_screenWidth, tmp_screenHeight);