Recherche avancée

Sur d’autres sites (13105)

• Transcoding & displaying user-uploaded videos for cross-browser/platform compatibility

  5 avril 2012, par Jonathan Amend

  I want to display user-uploaded videos in high quality on different browsers/platforms. I think I have a pretty good start, but there are a few issues. I have at my disposal ffmpeg 0.10 (called from PHP) and jwPlayer 5.8 (licensed).

  Ideally I would like to support :

  • Browsers with Flash 9+
  • Modern Firefox/Chrome/Safari without Flash
  • iOS 4.1+
  • Android 2.2+

  My current transcoding commands are :

  $commands = array(
  
      'flv' => "/usr/bin/ffmpeg -y -i {$sourceFile} -b 500k -ar 22050 -ab 64 {$tmpFileName}.flv 2>&1",
  
      'webm' => "/usr/bin/ffmpeg -i {$sourceFile} -acodec libvorbis -ac 2 -ab 96k -ar 44100 -b 500k {$tmpFileName}.webm 2>&1",
  
      'mp4' => "/usr/bin/ffmpeg -i {$sourceFile} -acodec libfaac -ab 96k -vcodec libx264 -level 21 -refs 2 -vf 'scale=trunc(ih*a/2)*2:trunc(iw/a/2)*2' -b 500k -bt 1000k -threads 0 -f mp4 {$tmpFileName}.pre.mp4 2>&1",
  
      'jpg' => "/usr/bin/ffmpeg -y -i {$sourceFile} -ss 5 -vcodec mjpeg -vframes 1 -an -f rawvideo {$tmpFileName}.jpg 2>&1"
  
  );
  
  
  
  $fastStartCommand = "/usr/bin/qt-faststart {$tmpFileName}.pre.mp4 {$tmpFileName}.mp4");

  My current display code is this :

  <video controls="controls" width="{$width}" height="{$height}" poster="{$fileJPG}" style="width: {$width}px; height: {$height}px;">
  
      <source src="{$fileMP4}" type="video/mp4" data-ext="mp4"></source>
  
      <source src="{$fileWEBM}" type="video/webm" data-ext="webm"></source>
  
      <a href="http://stackoverflow.com/feeds/tag/{$fileMP4}" title="Play Video">
  
          <img border="0" src="http://stackoverflow.com/feeds/tag/{$fileJPG}" width='0' height='0' alt="Play Video" />
  
      </a>
  
  </video>
  
  <code class="echappe-js">&lt;script type=&quot;text/javascript&quot;&gt;<br />
  
  swfobject.embedSWF(<br />
  
      &amp;#39;/flash/jwPlayer.swf&amp;#39;,<br />
  
      &amp;#39;video&amp;#39;,<br />
  
      {$width},<br />
  
      {$height},<br />
  
      &amp;#39;9.0.0&amp;#39;,<br />
  
      &amp;#39;&amp;#39;,<br />
  
      {<br />
  
          file: &amp;#39;{$fileFLV}&amp;#39;,<br />
  
          width: {$width},<br />
  
          height: {$height},<br />
  
          provider: &amp;#39;video&amp;#39;,<br />
  
          stretching: &amp;#39;uniform&amp;#39;,<br />
  
          smoothing: &amp;#39;true&amp;#39;,<br />
  
          dock: &amp;#39;true&amp;#39;<br />
  
      },<br />
  
      {<br />
  
          menu: &amp;#39;false&amp;#39;,<br />
  
          allowfullscreen: &amp;#39;true&amp;#39;,<br />
  
          allowscriptaccess: &amp;#39;always&amp;#39;,<br />
  
          allownetworking: &amp;#39;always&amp;#39;,<br />
  
          wmode: &amp;#39;transparent&amp;#39;<br />
  
      }<br />
  
  );<br />
  
  &lt;/script&gt;
  
  

  The idea is to prefer the Flash player (we have many custom skins that we would like to keep using), then fall back to HTML 5 video, and if that fails, show the JPEG with a link to the mp4 file (that seems to work as a last-ditch effort for iOS 2/3 and old browsers so they can at least play the video by launching QuickTime).

  The main issues that I'm not sure how to solve right now are :

  • How can I show a higher quality video in Flash ? I tried using the mp4 video instead of the flv but it doesn't work and I only get sound, no video. I think Flash is supposed to support h.264 these days, but how ? And do I have to use a different transcoding method for older and newer versions of Flash ?
  • How can I transcode the video with better quality ? Right now I am using the same bitrate for all resolutions. Any chance there is a way for ffmpeg to adjust the bitrate automatically ? Or can I detect and match the resolution of the uploaded video to a list of resolution -> bitrates somehow ?

• unexpected end of file error in ffmpeg libavutils/common.h while compiling

  23 mars 2014, par Mayank Agarwal

  I am compiling following simple code in ffmpeg.

  #include "stdafx.h"
  
  #include <libavcodec></libavcodec>avcodec.h>
  
  #include <libavformat></libavformat>avformat.h>
  
  
  
  void main()
  
  {
  
  
  
      av_register_all();
  
  
  
  } 

  all the ffmpeg headers i have included in project properties.
  but i am getting the following error on compiling.

  Error 1 error C1004 : unexpected end-of-file found dev\include\libavutil\common.h 87 1

  Regards
  Mayank

• FFmpeg/libav : YUV420P to RGB conversion

  17 janvier 2014, par learner

  I am working with Video encoding-decoding based on online tutorials. In the encoding section, the dummy image created is in YUV420P format. I need it to be in RGB or BGR format. Any idea how to do this ??

   #include 
  
   #include 
  
   #include 
  
  
  
   #ifdef HAVE_AV_CONFIG_H
  
   #undef HAVE_AV_CONFIG_H
  
   #endif
  
  
  
   #include "libavcodec/avcodec.h"
  
   #include "libavutil/mathematics.h"
  
  
  
   #define INBUF_SIZE 4096
  
   #define AUDIO_INBUF_SIZE 20480
  
   #define AUDIO_REFILL_THRESH 4096
  
  
  
   /*
  
   * Video encoding example
  
   */
  
   static void video_encode_example(const char *filename)
  
   {
  
       AVCodec *codec;
  
       AVCodecContext *c = NULL;
  
       int i, out_size, size, x, y, outbuf_size;
  
       FILE *f;
  
       AVFrame *picture;
  
       uint8_t *outbuf, *picture_buf;
  
  
  
       printf("Video encoding\n");
  
  
  
       /* find the mpeg1 video encoder */
  
       codec = avcodec_find_encoder(CODEC_ID_MPEG1VIDEO);
  
       if (!codec)
  
       {
  
           fprintf(stderr, "codec not found\n");
  
           exit(1);
  
       }
  
  
  
       c = avcodec_alloc_context();
  
       picture = avcodec_alloc_frame();
  
  
  
       /* put sample parameters */
  
       c->bit_rate = 400000;
  
       /* resolution must be a multiple of two */
  
       c->width = 352;
  
       c->height = 288;
  
       /* frames per second */
  
       c->time_base= (AVRational){1,25};
  
       c->gop_size = 10; /* emit one intra frame every ten frames */
  
       c->max_b_frames=1;
  
       c->pix_fmt = PIX_FMT_YUV420P;
  
  
  
       /* open it */
  
       if (avcodec_open(c, codec) < 0)
  
       {
  
           fprintf(stderr, "could not open codec\n");
  
           exit(1);
  
       }
  
  
  
       f = fopen(filename, "wb");
  
       if (!f)
  
       {
  
           fprintf(stderr, "could not open file! %s\n", filename);
  
           exit(1);
  
       }
  
  
  
       /* alloc image and output buffer */
  
       outbuf_size = 100000;
  
       outbuf = malloc(outbuf_size);
  
       size = c->width * c->height;
  
       picture_buf = malloc((size * 3) / 2); /* size for YUV 420 */
  
  
  
       picture->data[0] = picture_buf;
  
       picture->data[1] = picture->data[0] + size;
  
       picture->data[2] = picture->data[1] + size / 4;
  
       picture->linesize[0] = c->width;
  
       picture->linesize[1] = c->width / 2;
  
       picture->linesize[2] = c->width / 2;
  
  
  
       /* encode 1 second of video */
  
       for(i=0; i<25; i++)
  
       {
  
           fflush(stdout);
  
           /* prepare a dummy image */
  
           /* Y */
  
           for(y=0; y < c->height; y++)
  
           {
  
               for(x=0; x < c->width; x++)
  
               {
  
                   picture->data[0][y * picture->linesize[0] + x] = x + y + i * 3;
  
               }
  
           }
  
  
  
           /* Cb and Cr */
  
           for(y=0; y < c->height/2; y++)
  
           {
  
               for(x=0; x < c->width/2; x++)
  
               {
  
                   picture->data[1][y * picture->linesize[1] + x] = 128 + y + i * 2;
  
                   picture->data[2][y * picture->linesize[2] + x] = 64 + x + i * 5;
  
               }
  
           }
  
  
  
           /* encode the image */
  
           out_size = avcodec_encode_video(c, outbuf, outbuf_size, picture);
  
           printf("encoding frame %3d (size=%5d)\n", i, out_size);
  
           fwrite(outbuf, 1, out_size, f);
  
       }
  
  
  
       /* get the delayed frames */
  
       for(; out_size; i++)
  
       {
  
           fflush(stdout);
  
  
  
           out_size = avcodec_encode_video(c, outbuf, outbuf_size, NULL);
  
           printf("write frame %3d (size=%5d)\n", i, out_size);
  
           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);
  
       free(picture_buf);
  
       free(outbuf);
  
  
  
       avcodec_close(c);
  
       av_free(c);
  
       av_free(picture);
  
       printf("\n");
  
   }
  
  
  
   /*
  
    * Video decoding example
  
    */
  
  
  
   static void pgm_save(unsigned char *buf, int wrap, int xsize, int ysize,
  
                        char *filename)
  
   {
  
       FILE *f;
  
       int i;
  
  
  
       f=fopen(filename,"w");
  
       fprintf(f,"P5\n%d %d\n%d\n",xsize,ysize,255);
  
       for(i=0; i* set end of buffer to 0 (this ensures that no overreading happens for damaged mpeg streams) */
  
       memset(inbuf + INBUF_SIZE, 0, FF_INPUT_BUFFER_PADDING_SIZE);
  
  
  
       printf("Video decoding\n");
  
  
  
       /* find the mpeg1 video decoder */
  
       codec = avcodec_find_decoder(CODEC_ID_MPEG1VIDEO);
  
       if (!codec) {
  
           fprintf(stderr, "codec not found\n");
  
           exit(1);
  
       }
  
  
  
       c = avcodec_alloc_context();
  
       picture = avcodec_alloc_frame();
  
  
  
       if(codec->capabilities&CODEC_CAP_TRUNCATED)
  
           c->flags|= CODEC_FLAG_TRUNCATED; /* we do not send complete frames */
  
  
  
       /* For some codecs, such as msmpeg4 and mpeg4, width and height
  
          MUST be initialized there because this information is not
  
          available in the bitstream. */
  
  
  
       /* open it */
  
       if (avcodec_open(c, codec) < 0) {
  
           fprintf(stderr, "could not open codec\n");
  
           exit(1);
  
       }
  
  
  
       /* the codec gives us the frame size, in samples */
  
  
  
       f = fopen(filename, "rb");
  
       if (!f) {
  
           fprintf(stderr, "could not open file! %s\n", filename);
  
           exit(1);
  
       }
  
  
  
       frame = 0;
  
       for(;;) {
  
           avpkt.size = fread(inbuf, 1, INBUF_SIZE, f);
  
           if (avpkt.size == 0)
  
               break;
  
  
  
           /* NOTE1: some codecs are stream based (mpegvideo, mpegaudio)
  
              and this is the only method to use them because you cannot
  
              know the compressed data size before analysing it.
  
  
  
              BUT some other codecs (msmpeg4, mpeg4) are inherently frame
  
              based, so you must call them with all the data for one
  
              frame exactly. You must also initialize 'width' and
  
              'height' before initializing them. */
  
  
  
           /* NOTE2: some codecs allow the raw parameters (frame size,
  
              sample rate) to be changed at any frame. We handle this, so
  
              you should also take care of it */
  
  
  
           /* here, we use a stream based decoder (mpeg1video), so we
  
              feed decoder and see if it could decode a frame */
  
  
  
           avpkt.data = inbuf;
  
           while (avpkt.size > 0)
  
           {
  
               len = avcodec_decode_video2(c, picture, &got_picture, &avpkt);
  
               if (len < 0)
  
               {
  
                   fprintf(stderr, "Error while decoding frame %d\n", frame);
  
                   exit(1);
  
               }
  
               if (got_picture)
  
               {
  
                   printf("saving frame %3d\n", frame);
  
                   fflush(stdout);
  
  
  
                   /* the picture is allocated by the decoder. no need to
  
                      free it */
  
                   snprintf(buf, sizeof(buf), outfilename, frame);
  
                   pgm_save(picture->data[0], picture->linesize[0],
  
                            c->width, c->height, buf);
  
                   frame++;
  
               }
  
               avpkt.size -= len;
  
               avpkt.data += len;
  
           }
  
       }
  
  
  
       /* some codecs, such as MPEG, transmit the I and P frame with a
  
          latency of one frame. You must do the following to have a
  
          chance to get the last frame of the video */
  
  
  
       avpkt.data = NULL;
  
       avpkt.size = 0;
  
       len = avcodec_decode_video2(c, picture, &got_picture, &avpkt);
  
       if (got_picture)
  
       {
  
           printf("saving last frame %3d\n", frame);
  
           fflush(stdout);
  
  
  
           /* the picture is allocated by the decoder. no need to
  
              free it */
  
           snprintf(buf, sizeof(buf), outfilename, frame);
  
           pgm_save(picture->data[0], picture->linesize[0],
  
                    c->width, c->height, buf);
  
           frame++;
  
       }
  
  
  
       fclose(f);
  
  
  
       avcodec_close(c);
  
       av_free(c);
  
       av_free(picture);
  
       printf("\n");
  
   }
  
  
  
   int main(int argc, char **argv)
  
   {
  
       const char *filename;
  
  
  
       /* must be called before using avcodec lib */
  
       avcodec_init();
  
  
  
       /* register all the codecs */
  
       avcodec_register_all();
  
  
  
       if (argc <= 1)
  
       { 
  
           video_encode_example("asdf.mpg");
  
           filename = "asdf.mpg";
  
       }
  
       else
  
       {
  
           filename = argv[1];
  
       }
  
  
  
       video_decode_example("%d.pgm", filename);
  
  
  
       return 0;
  
  }

  I tried with PIX_FMT_RGB24 and changing each channel accordingly. For this, I declared another :

  AVFrame *pictureRGB ; and then : pictureRGB = avcodec_alloc_frame() ;

  for(y=0; y < c->height; y++)
  
  {
  
       for(x=0; x < c->width; x++)
  
       {
  
            pictureRGB ->data[0][y * pictureRGB ->linesize[0] + x] = x + y + i * 3;
  
            pictureRGB ->data[1][y * pictureRGB ->linesize[1] + x] = x + y + i * 3;
  
            pictureRGB ->data[2][y * pictureRGB ->linesize[2] + x] = x + y + i * 3;
  
       }
  
  }

  But it gives errors ! I am totally new to this library. Is it possible to directly encode as a RGB dummy image rather than as YUV420P. Anybody out there with sound knowledge in this area ??? Thanks in advance !