Recherche avancée

Sur d’autres sites (5476)

• Swscale - image patch (NV12) color conversion - invalid border

  6 juin 2019, par dajuric

  The goal is to convert NV12 to BGR24 image, more exactly an image patch (x:0, y:0, w:220, h:220).
  The issue is the undefined pixel column on the right of the converted patch as shown :
  

  The question is why is this happening (even though the coordinates and the dimensions of the patch have even values) ?
  (Interestingly enough for an odd width value, that issue is not present)

  ---

  The patch has the following bounding box : (x:0, y:0, w:220, h:220).
  The behavior should be reproducible with any image. Conversion can be done using the ppm conversion page.

  The following code creates a nv12 image from a bgr24 image and then converts a nv12 patch back to bgr24 patch. If everything worked properly the output should have been identical to a source image.

  #include <libswscale></libswscale>swscale.h>
  
  #include <libavutil></libavutil>imgutils.h>
  
  
  
  void readPPM(const char* filename, uint8_t** bgrData, int* stride, int* w, int* h)
  
  {
  
      FILE* fp = fopen(filename, "rb");
  
      fscanf(fp, "%*s\n"); //skip format check
  
  
  
      fscanf(fp, "%d %d\n", w, h);
  
      fscanf(fp, "%*d\n"); //skip max value check
  
  
  
      *stride = *w * 3;
  
      *bgrData = av_malloc(*h * *stride);
  
  
  
      for (int r = 0; r &lt; *h; r++)
  
      {
  
          uint8_t* rowData = *bgrData + r * *stride;
  
          for (int c = 0; c &lt; *w; c++)
  
          {
  
              //rgb -> bgr
  
              fread(&amp;rowData[2], 1, 1, fp);
  
              fread(&amp;rowData[1], 1, 1, fp);
  
              fread(&amp;rowData[0], 1, 1, fp);
  
  
  
              rowData += 3;
  
          }
  
      }
  
  
  
      fclose(fp);
  
  }
  
  
  
  void writePPM(const char* filename, uint8_t* bgrData, int stride, int w, int h)
  
  {
  
      FILE* fp = fopen(filename, "wb");
  
      fprintf(fp, "P6\n");
  
      fprintf(fp, "%d %d\n", w, h);
  
      fprintf(fp, "%d\n", 255);
  
  
  
      for (int r = 0; r &lt; h; r++)
  
      {
  
          uint8_t* rowData = bgrData + r * stride;
  
          for (int c = 0; c &lt; w; c++)
  
          {
  
              //bgr -> rgb
  
              fwrite(&amp;rowData[2], 1, 1, fp);
  
              fwrite(&amp;rowData[1], 1, 1, fp);
  
              fwrite(&amp;rowData[0], 1, 1, fp);
  
  
  
              rowData += 3;       
  
          }
  
      }
  
  
  
      fclose(fp);
  
  }
  
  
  
  
  
  void bgrToNV12(uint8_t* srcData[4], int srcStride[4], 
  
                 uint8_t* tgtData[4], int tgtStride[4],
  
                 int w, int h)
  
  {
  
      struct SwsContext* context = sws_getContext(w, h, AV_PIX_FMT_BGR24,
  
                                                  w, h, AV_PIX_FMT_NV12, SWS_POINT, NULL, NULL, NULL);
  
      {
  
          sws_scale(context,
  
                    srcData, srcStride, 0, h,
  
                    tgtData, tgtStride);
  
      }
  
      sws_freeContext(context);
  
  }
  
  
  
  void nv12ToBgr(uint8_t* srcData[4], int srcStride[4],
  
                 uint8_t* tgtData[4], int tgtStride[4],
  
                 int w, int h)
  
  {
  
      struct SwsContext* context = sws_getContext(w, h, AV_PIX_FMT_NV12,
  
                                                  w, h, AV_PIX_FMT_BGR24, SWS_POINT, NULL, NULL, NULL);
  
      {
  
          sws_scale(context,
  
                    srcData, srcStride, 0, h,
  
                    tgtData, tgtStride);
  
      }
  
      sws_freeContext(context);
  
  }
  
  
  
  
  
  int main()
  
  {
  
      //load BGR image
  
      uint8_t* bgrData[4]; int bgrStride[4]; int bgrW, bgrH;
  
      readPPM("sample.ppm", &amp;bgrData[0], &amp;bgrStride[0], &amp;bgrW, &amp;bgrH);
  
  
  
      //create NV12 image from the BGR image
  
      uint8_t* nv12Data[4]; int nv12Stride[4];
  
      av_image_alloc(nv12Data, nv12Stride, bgrW, bgrH, AV_PIX_FMT_NV12, 16);
  
      bgrToNV12(bgrData, bgrStride, nv12Data, nv12Stride, bgrW, bgrH);
  
  
  
      //convert nv12 patch to bgr patch
  
      nv12ToBgr(nv12Data, nv12Stride, bgrData, bgrStride, 220, 220);   //invalid result (random column stripe)
  
      //nv12ToBgr(nv12Data, nv12Stride, bgrData, bgrStride, 221, 220); //valid result
  
  
  
      //save bgr image (should be exactly as original BGR image)
  
      writePPM("sample-out.ppm", bgrData[0], bgrStride[0], bgrW, bgrH);
  
  
  
      //cleanup
  
      av_freep(bgrData);
  
      av_freep(nv12Data);
  
      return 0;
  
  }

• Making a ffmpeg screen capture on Mac OS X using YUV 4:2:0 Planar color model

  30 mai 2019, par Bass

  I make screen recordings with ffmpeg, using avfoundation on Mac OS X, x11grab on Linux and gdigrab on Windows.

  The resulting files should be compatible with modern web browsers (<video></video>), so I use H.264 codec and request YUV 4:2:0 Planar pixel format.

  On Mac OS X, however (unlike Linux and Windows), I receive the following logging :

  /usr/local/bin/ffmpeg -y -v error -f avfoundation -threads 0 -hide_banner -i 1:none -f mp4 -vcodec h264 -pix_fmt yuv420p -r 25/1 -qscale:v 1 -vf scale=-1:1080 target.mp4
  
  [avfoundation @ 0x7fdba2003a00] Selected pixel format (yuv420p) is not supported by the input device.
  
  [avfoundation @ 0x7fdba2003a00] Supported pixel formats:
  
  [avfoundation @ 0x7fdba2003a00]   uyvy422
  
  [avfoundation @ 0x7fdba2003a00]   yuyv422
  
  [avfoundation @ 0x7fdba2003a00]   nv12
  
  [avfoundation @ 0x7fdba2003a00]   0rgb
  
  [avfoundation @ 0x7fdba2003a00]   bgr0

  Still, according to mplayer, the resulting MP4 file seems to have YUV 4:2:0 Planar color model :

  [h264 @ 0x1048a8ac0]Format yuv420p chosen by get_format().
  
  [h264 @ 0x1048a8ac0]Reinit context to 1728x1088, pix_fmt: yuv420p
  
  [h264 @ 0x1048a8ac0]Format yuv420p chosen by get_format().
  
  [h264 @ 0x1048a8ac0]Reinit context to 1728x1088, pix_fmt: yuv420p
  
  [swscaler @ 0x1048c3cc0]bicubic scaler, from yuv420p to yuyv422 using MMXEXT
  
  *** [scale] Exporting mp_image_t, 1728x1080x12bpp YUV planar, 2799360 bytes
  
  *** [vo] Allocating mp_image_t, 1728x1080x16bpp YUV packed, 3732480 bytes

  the same confirmed by ffmpeg :

  $ ffmpeg -i target.mp4 -hide_banner
  
  Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'target.mp4':
  
    Metadata:
  
      major_brand     : isom
  
      minor_version   : 512
  
      compatible_brands: isomiso2avc1mp41
  
      encoder         : Lavf58.20.100
  
    Duration: 00:00:04.72, start: 0.000000, bitrate: 201 kb/s
  
      Stream #0:0(und): Video: h264 (High) (avc1 / 0x31637661), yuv420p, 1728x1080, 197 kb/s, 25 fps, 25 tbr, 12800 tbn, 50 tbc (default)
  
      Metadata:
  
        handler_name    : VideoHandler

  Questions :

  1. Can someone explain the above ffmpeg logging ?
  2. If I still need to convert the avfoundation video stream to yuv420p, how do I make it on the fly (in a single ffmpeg pass) ?

• create video with fluctuated background color and drawbox with fluctuated color ?

  27 mai 2019, par Wang

  Is there anyway to create a video with fluctuated background color choosing from a list for example [R, G, B] the background color along time will be R,G,B,R,G,B .... I might use nullsrc with geq. But seems the geq is quite slow which apply to each pixel.

  I would also like to change the color of drawbox too, is there any easy way to do it ? I knew we might be able to use openclsrc but it seems overkill for this simple task.

  Update, for drawbox I found a way, which is kind of abuse the timeline editing :

  ffmpeg -f lavfi -i smptebars=r=30:d=30:size=800x600 -vf "drawbox=enable='eq(mod(n,3),0)':x=0:y=0:t=20:c=red,drawbox=enable='eq(mod(n,3),1)':x=0:y=0:t=20:c=green,drawbox=enable='eq(mod(n,3),2)':x=0:y=0:t=20:c=blue" /tmp/test_box.mp4