Newest 'ffmpeg' Questions - Stack Overflow

Les articles publiés sur le site

• How to reconnect using avformat_open_input without having to alloc the decoder again ?

  26 septembre 2012, par Jona

  Currently, I have a code based on ffplay to stream live content.

  One thing I'm looking to do is be able to reconnect upon loosing a connection without having to shutdown the whole decoding process.

  To me the solution is alloc the decoder myself once and keep using it across reconnections. I can't seem to figure out how to setup a decoder without having to depend on the AVFormatContext. Right now my code is failing when trying to use my own allocated AVCodecContext to decode. But it doesn't fail if I use the AVCodecContext given by AVFormatContext.

  This is part of my initial code:

  // attempt to find more information about the codec
  // also it will open the codecs needed.
  fferror = avformat_find_stream_info(ic, NULL);
  if (0 > fferror)
  {
      // TODO verify type of error to better map it to our errors
      error = ERROR_FAIL_TO_CONNECT;
      LOGE("download() -> avformat_find_stream_info failed! fferror:%d, error:%d", fferror, error);
      goto fail;
  }
  
  AVCodec *dec;
  // select the audio stream
  int ret = av_find_best_stream(ic, AVMEDIA_TYPE_AUDIO, -1, -1, &dec, 0);
  if (0 > ret) {
      error = ERROR_UNEXPECTED_ERROR;
      LOGE("download() -> av_find_best_stream failed! ret:%d, error:%d", ret, error);
      goto fail;
  }
  
  LOGI("download() -> STREAM: nb_streams:%d", ic->nb_streams);
  LOGI("download() -> STREAM: audio format:%s", ic->iformat->name);
  LOGI("download() -> STREAM: audio bitrate:%d", ic->bit_rate);
  
  
  // save the audio stream index and source
  is->audio_stream_index = ret;
  is->audio_st = ic->streams[is->audio_stream_index];
  is->audio_buf_size = 0;
  is->audio_buf_index = 0;
  
  is->audio_st->discard = AVDISCARD_DEFAULT;
  
  if(ic->pb) {
      ic->pb->eof_reached= 0; //FIXME hack, ffplay maybe should not use url_feof() to test for the end
  }
  
  if (show_status) {
          av_dump_format(ic, 0, is->filename, 0);
      }
  
  // open codec
      error = open_decoder(is->audio_st->codec, dec);
      if (ERROR_NO_ERROR != error) {
          LOGE("receive_thread() -> open_decoder failed! error:%d", error);
          goto fail;
      }
  

  And this is the funtion that initializes the decoder.

  static int open_decoder (AVCodecContext *avctx, AVCodec *codec)
  {
      int fferror         = 0;
      AVCodecContext *c   = NULL;
  
      if (smDecoder.open) {
          LOGW("open_decoder() -> decoder is already open!");
          return ERROR_NO_ERROR;
      }
  
      // find the decoder
      if (!codec)
      {
          codec = avcodec_find_decoder(avctx->codec_id);
          if (!codec)
          {
              LOGE("open_decoder() -> avcodec_find_decoder failed!");
              return ERROR_UNEXPECTED_ERROR;
          }
      }
  
      // allocate the decoder av context
      c = avcodec_alloc_context3(codec);
      if (NULL == c) {
          LOGE("open_decoder() -> avcodec_alloc_context3 failed! Out of memory?");
          return ERROR_UNEXPECTED_ERROR;
      }
  
      // check if the type of codec we support
      if (AVMEDIA_TYPE_AUDIO != c->codec_type)
      {
          LOGE("open_decoder() -> codec_type not supported! codec_type:%d",c->codec_type);
          return ERROR_UNEXPECTED_ERROR;
      }
  
      // set the proper channels if not properly set
      if (c->channels > 0) {
          c->request_channels = FFMIN(2, c->channels);
      } else {
          c->request_channels = 2;
      }
  
      c->debug_mv = 0;
      c->debug = 0;
      c->workaround_bugs = workaround_bugs;
      c->idct_algo= idct;
      if(fast) c->flags2 |= CODEC_FLAG2_FAST;
      c->error_concealment= error_concealment;
      c->thread_count = thread_count;
  
      // open the decoder
      fferror = avcodec_open2(avctx, codec, NULL);
      if (fferror < 0)
      {
          LOGE("open_decoder() -> avcodec_open2 failed! fferror:%d", fferror);
          return ERROR_UNEXPECTED_ERROR;
      }
  
      // clean up our reusable packet
      memset(&smDecoder.audio_pkt, 0, sizeof(smDecoder.audio_pkt));
  
      smDecoder.open = 1;
      smDecoder.codec = codec;
      smDecoder.avctx = c;
  
      return ERROR_NO_ERROR;
  }
  

• ffmpeg result to a tempfile

  26 septembre 2012, par user1165201

  I am new to python and ffmpeg. I have a following question to ask.

  If I run the following command from command-line and it works.

  ffmpeg -i  1.flv  temp_filename
  

  If I put it in a program

     temp_file_handle, temp_filename = tempfile.mkstemp('.flv')
  
     command = "ffmpeg -i " + newvideo.location + " "+ temp_filename
  
     out = subprocess.Popen(command, stdout=subprocess.PIPE, stderr=subprocess.PIPE, shell=True)
     res = out.communicate()
  

  The generated video didn't write to the tem_filename. Why?

• ffmpeg keyframe extraction

  26 septembre 2012, par Shikhar Shrivastav

  I have been trying to extract keyframes from video using ffmpeg 0.11.1 . So far all the commands I have tried do not extract keyframes but return all the frames ie 25fps*total time number of frames in the output. I tried setting the keyint_min as 25 to make sure there is a amximum of 1 keyframe per second.

  ffmpeg -vf select="eq(pict_type\,PICT_TYPE_I)" -g 250 -keyint_min 25 -i C:\test.mp4 -vsync 2 -f image2 C:\testTemp\thumbnails-%02d.jpeg
  

  But still all the frames are returned.

  Then i tried, to separate the keyframes by 20 seconds.

  ffmpeg -i C:\test.mp4 -vf select='eq(pict_type\,I)*(isnan(prev_selected_t)+gte(t-prev_selected_t\,20))' -vsync 0 -f image2 C:\testTemp\%09d.jpg
  

  Again same result, all the frames are returned.

  What should I do?

• Encode image to video using ffmpeg (sws_scale)

  26 septembre 2012, par bahar_p

  I'm trying to encode an image to video using ffmpeg library. I have these global params:

  //Global params
  AVCodec         *codec;
  AVCodecContext  *codecCtx;
  uint8_t         *output_buffer;
  int             output_buffer_size;
  

  I divided the encoding to 3 methods: Initialize the encoder:

  jint Java_com_camera_simpledoublewebcams2_CameraPreview_initencoder(JNIEnv* env,jobject thiz){
  avcodec_register_all();
  avcodec_init();
  av_register_all();
  
  int fps = 30;
  
  /* find the H263 video encoder */
  codec = avcodec_find_encoder(CODEC_ID_H263);
  if (!codec) {
      LOGI("avcodec_find_encoder() run fail.");
      return -5;
  }
  
  //allocate context
  codecCtx = avcodec_alloc_context();
  
  /* put sample parameters */
  codecCtx->bit_rate = 400000;
  /* resolution must be a multiple of two */
  codecCtx->width = 176;
  codecCtx->height = 144;
  /* frames per second */
  codecCtx->time_base = (AVRational){1,fps};
  codecCtx->pix_fmt = PIX_FMT_YUV420P;
  codecCtx->codec_id = CODEC_ID_H263;
  codecCtx->codec_type = AVMEDIA_TYPE_VIDEO;
  
  /* open it */
  if (avcodec_open(codecCtx, codec) < 0) {
      LOGI("avcodec_open() run fail.");
      return -10;
  }
  
  //init buffer
  output_buffer_size = 500000;
  output_buffer = malloc(output_buffer_size);
  
  return 0;
  

  }

  Encoding the image:

  jint Java_com_camera_simpledoublewebcams2_CameraPreview_encodejpeg(JNIEnv* env,jobject thiz,jchar* cImage, jint imageSize){
  int             out_size;
  AVFrame         *picture;
  AVFrame         *outpic;
  uint8_t         *outbuffer;
  
  //allocate frame    
  picture = avcodec_alloc_frame();    
  outpic = avcodec_alloc_frame();
  
  int nbytes = avpicture_get_size(PIX_FMT_YUV420P, codecCtx->width, codecCtx->height);
  outbuffer = (uint8_t*)av_malloc(nbytes);
  outpic->pts = 0;
  
  //fill picture with image
  avpicture_fill((AVPicture*)picture, (uint8_t*)cImage, PIX_FMT_RGBA, codecCtx->width, codecCtx->height);
  //fill outpic with empty image
  avpicture_fill((AVPicture*)outpic, outbuffer, PIX_FMT_YUV420P, codecCtx->width, codecCtx->height);
  
  //rescale the image 
  struct SwsContext* fooContext = sws_getContext(codecCtx->width, codecCtx->height, 
                                                         PIX_FMT_RGBA, 
                                                         codecCtx->width, codecCtx->height, 
                                                         PIX_FMT_YUV420P, 
                                                         SWS_FAST_BILINEAR, NULL, NULL, NULL);
  sws_scale(fooContext, picture->data, picture->linesize, 0, codecCtx->height, outpic->data, outpic->linesize);   
  
  //encode the image
  out_size = avcodec_encode_video(codecCtx, output_buffer, output_buffer_size, outpic);
  out_size += avcodec_encode_video(codecCtx, output_buffer, output_buffer_size, outpic);
  
  //release pictures
  av_free(outbuffer);
  av_free(picture);
  av_free(outpic);
  
  return out_size;
  

  }

  And closing the encoder:

  void Java_com_camera_simpledoublewebcams2_CameraPreview_closeencoder(JNIEnv* env,jobject thiz){
  free(output_buffer);
  avcodec_close(codecCtx);
  av_free(codecCtx);
  

  }

  When I send the first image, I get a result from the encoder. When I try to send another image the program crashes. I tried calling init once and then the images, then the close - didn't work. I tried calling the init and the close for every image - didn't work.

  Any suggestions?

  Thanks!

  EDIT: After further research I found that the problem is at sws_scale method. Still don't know what is causing this issue...

• Compiling ffmpeg for use in Android (ndk) for x86 architecture

  26 septembre 2012, par Ankur22

  I am trying to compile ffmpeg for Android but for the x86 architecture. I've successfully compiled an arm .so and it all works, and I though that by replacing arm with x86 in the relevant places within the build script would have done the trick. Unfortunately that's not the case and I get some strange things occurring.

  First I get this while it's compiling ffmpeg:

  /home/ankur/android-ndk-r8/platforms/android-8/arch-arm//usr/include/strings.h:49: warning: redundant redeclaration of 'index'
  

  Notice the arch-arm instead of arch-x86. Finally I get this:

  /home/ankur/android-ndk-r8/toolchains/x86-4.4.3/prebuilt/linux-x86/bin/i686-android-linux-ld: libavcodec/libavcodec.a(4xm.o): Relocations in generic ELF (EM: 40)
  /home/ankur/android-ndk-r8/toolchains/x86-4.4.3/prebuilt/linux-x86/bin/i686-android-linux-ld: libavcodec/libavcodec.a(4xm.o): Relocations in generic ELF (EM: 40)
  /home/ankur/android-ndk-r8/toolchains/x86-4.4.3/prebuilt/linux-x86/bin/i686-android-linux-ld: libavcodec/libavcodec.a(4xm.o): Relocations in generic ELF (EM: 40)
  libavcodec/libavcodec.a(4xm.o): could not read symbols: File in wrong format
  

  The build script I've created looks like this:

  #!/bin/bash
  
  NDK=~/android-ndk-r8
  PLATFORM=$NDK/platforms/android-8/arch-x86/
  PREBUILT=$NDK/toolchains/x86-4.4.3/prebuilt/linux-x86
  function build_one_r8
  {
  ./configure \
      --disable-shared \
      --enable-static \
      --enable-gpl \
      --enable-version3 \
      --enable-nonfree \
      --disable-doc \
      --disable-ffmpeg \
      --disable-ffplay \
      --disable-ffprobe \
      --disable-ffserver \
      --disable-avdevice \
      --disable-avfilter \
      --disable-postproc \
      --enable-small \
      --cross-prefix=$PREBUILT/bin/i686-android-linux- \
      --enable-cross-compile \
      --target-os=linux \
      --extra-cflags="-I$PLATFORM/usr/include" \
      --arch=x86 \
      --disable-symver \
      --disable-debug \
      --disable-stripping \
      $ADDITIONAL_CONFIGURE_FLAG
  sed -i 's/HAVE_LRINT 0/HAVE_LRINT 1/g' config.h
  sed -i 's/HAVE_LRINTF 0/HAVE_LRINTF 1/g' config.h
  sed -i 's/HAVE_ROUND 0/HAVE_ROUND 1/g' config.h
  sed -i 's/HAVE_ROUNDF 0/HAVE_ROUNDF 1/g' config.h
  sed -i 's/HAVE_TRUNC 0/HAVE_TRUNC 1/g' config.h
  sed -i 's/HAVE_TRUNCF 0/HAVE_TRUNCF 1/g' config.h
  make clean
  make  -j4 install
  $PREBUILT/bin/i686-android-linux-ar d libavcodec/libavcodec.a inverse.o
  $PREBUILT/bin/i686-android-linux-ld -rpath-link=$PLATFORM/usr/lib -L$PLATFORM/usr/lib  -soname libffmpeg.so -shared -nostdlib  -z,noexecstack -Bsymbolic --whole-archive --no-undefined -o $PREFIX/libffmpeg.so libavcodec/libavcodec.a libavformat/libavformat.a libavutil/libavutil.a libswscale/libswscale.a -lc -lm -lz -ldl -llog  --warn-once  --dynamic-linker=/system/bin/linker $PREBUILT/lib/gcc/i686-android-linux/4.4.3/libgcc.a
  }
  function build_one_r8_2
  {
  $PREBUILT/bin/i686-android-linux-ar d libavcodec/libavcodec.a inverse.o
  $PREBUILT/bin/i686-android-linux-ld -rpath-link=$PLATFORM/usr/lib -L$PLATFORM/usr/lib  -soname libffmpeg.so -shared -nostdlib  -z,noexecstack -Bsymbolic --whole-archive --no-undefined -o $PREFIX/libffmpeg.so libavcodec/libavcodec.a libavformat/libavformat.a libavutil/libavutil.a libswscale/libswscale.a -lc -lm -lz -ldl -llog  --warn-once  --dynamic-linker=/system/bin/linker $PREBUILT/lib/gcc/i686-android-linux/4.4.3/libgcc.a
  }
  #x86
  CPU=x86
  OPTIMIZE_CFLAGS="-march=$CPU "
  PREFIX=./android/$CPU
  ADDITIONAL_CONFIGURE_FLAG=
  build_one_r8
  

  I hope I'm doing something wrong rather than this not being possible.

  Thanks!