Recherche avancée

Sur d’autres sites (10989)

• Image to MPEG on Linux works, same code on Android = green video

  5 avril 2018, par JScoobyCed

  EDIT
  I have check the execution and found that the error is not (yet) at the swscale point. My current issue is that the JPG image is not found :
  No such file or directory
  when doing the avformat_open_input(&pFormatCtx, imageFileName, NULL, NULL);
  Before you tell me I need to register anything, I can tell I already did (I updated the code below).
  I also added the Android permission to access the external storage (I don’t think it is related to Android since I can already write to the /mnt/sdcard/ where the image is also located)
  END EDIT

  I have been through several tutorials (including the few posted from SO, i.e. http://dranger.com/ffmpeg/, how to compile ffmpeg for Android...,been through dolphin-player source code). Here is what I have :
  . Compiled ffmpeg for android
  . Ran basic tutorials using NDK to create a dummy video on my android device
  . been able to generate a MPEG2 video from images on Ubuntu using a modified version of dummy video code above and a lot of Googling
  . running the new code on Android device gives a green screen video (duration 1 sec whatever the number of frames I encode)

  I saw another post about iPhone in a similar situation that mentioned the ARM processor optimization could be the culprit. I tried a few ldextra-flags (-arch armv7-a and similar) to no success.

  I include at the end the code that loads the image. Is there something different to do on Android than on linux ? Is my ffmpeg build not correct for Android video encoding ?

  void copyFrame(AVCodecContext *destContext, AVFrame* dest,
  
              AVCodecContext *srcContext, AVFrame* source) {
  
  struct SwsContext *swsContext;
  
  swsContext = sws_getContext(srcContext->width, srcContext->height, srcContext->pix_fmt,
  
                  destContext->width, destContext->height, destContext->pix_fmt,
  
                  SWS_FAST_BILINEAR, NULL, NULL, NULL);
  
  sws_scale(swsContext, source->data, source->linesize, 0, srcContext->height, dest->data, dest->linesize);
  
  sws_freeContext(swsContext);
  
  }
  
  
  
  int loadFromFile(const char* imageFileName, AVFrame* realPicture, AVCodecContext* videoContext) {
  
  AVFormatContext *pFormatCtx = NULL;
  
  avcodec_register_all();
  
  av_register_all();
  
  
  
  int ret = avformat_open_input(&pFormatCtx, imageFileName, NULL, NULL);
  
  if (ret != 0) {
  
      // ERROR hapening here
  
      // Can't open image file. Use strerror(AVERROR(ret))) for details
  
      return ERR_CANNOT_OPEN_IMAGE;
  
  }
  
  
  
  AVCodecContext *pCodecCtx;
  
  
  
  pCodecCtx = pFormatCtx->streams[0]->codec;
  
  pCodecCtx->width = W_VIDEO;
  
  pCodecCtx->height = H_VIDEO;
  
  pCodecCtx->pix_fmt = PIX_FMT_YUV420P;
  
  
  
  // Find the decoder for the video stream
  
  AVCodec *pCodec = avcodec_find_decoder(pCodecCtx->codec_id);
  
  if (!pCodec) {
  
      // Codec not found
  
      return ERR_CODEC_NOT_FOUND;
  
  }
  
  
  
  // Open codec
  
  if (avcodec_open(pCodecCtx, pCodec) < 0) {
  
      // Could not open codec
  
      return ERR_CANNOT_OPEN_CODEC;
  
  }
  
  
  
  // 
  
  AVFrame *pFrame;
  
  
  
  pFrame = avcodec_alloc_frame();
  
  
  
  if (!pFrame) {
  
      // Can't allocate memory for AVFrame
  
      return ERR_CANNOT_ALLOC_MEM;
  
  }
  
  
  
  int frameFinished;
  
  int numBytes;
  
  
  
  // Determine required buffer size and allocate buffer
  
  numBytes = avpicture_get_size(PIX_FMT_YUV420P, pCodecCtx->width, pCodecCtx->height);
  
  uint8_t *buffer = (uint8_t *) av_malloc(numBytes * sizeof (uint8_t));
  
  
  
  avpicture_fill((AVPicture *) pFrame, buffer, PIX_FMT_YUV420P, pCodecCtx->width, pCodecCtx->height);
  
  AVPacket packet;
  
  int res = 0;
  
  while (av_read_frame(pFormatCtx, &packet) >= 0) {
  
      if (packet.stream_index != 0)
  
          continue;
  
  
  
      ret = avcodec_decode_video2(pCodecCtx, pFrame, &frameFinished, &packet);
  
      if (ret > 0) {
  
          // now, load the useful info into realPicture
  
          copyFrame(videoContext, realPicture, pCodecCtx, pFrame);
  
          // Free the packet that was allocated by av_read_frame
  
          av_free_packet(&packet);
  
          return 0;
  
      } else {
  
          // Error decoding frame. Use strerror(AVERROR(ret))) for details
  
          res = ERR_DECODE_FRAME;
  
      }
  
  }
  
  av_free(pFrame);
  
  
  
  // close codec
  
  avcodec_close(pCodecCtx);
  
  
  
  // Close the image file
  
  av_close_input_file(pFormatCtx);
  
  
  
  return res;
  
  }

  Some ./configure options :
  --extra-cflags="-O3 -fpic -DANDROID -DHAVE_SYS_UIO_H=1 -Dipv6mr_interface=ipv6mr_ifindex -fasm -Wno-psabi -fno-short-enums -fno-strict-aliasing -finline-limit=300 -mfloat-abi=softfp -mfpu=vfp -marm -march=armv7-a -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE"

  --extra-ldflags="-Wl,-rpath-link=$PLATFORM/usr/lib -L$PLATFORM/usr/lib -nostdlib -lc -lm -ldl -llog"

  --arch=armv7-a --enable-armv5te --enable-armv6 --enable-armvfp --enable-memalign-hack

• ffmpeg code not working aws lambda environment

  5 janvier 2021, par Akash Das

  I followed this article and deployed the Lambda functions and layers and set it up all fine.

  


  But I am facing a problem when I try converting the mp4 video into a 720p video, the ffmpeg commands that work in my Ubuntu 20.04 are not working in the Lambda and gives me error in the form of 0kb files in the destination folder.

  


  Can someone explain why this is happening ? The command that I used was :

  


  ffmpeg -i input.mp4 -s 1280x720 -c:a copy output.mp4


  


  I use this command in the Python file

  


  ffmpeg_cmd = "/opt/bin/ffmpeg -i \"" + s3_source_signed_url + "\" -f mp4 -s 1280x720 -c:a copy -" ___ . 


  


  This is the error I get :

  


  

  


  The problem is that I used a code like this to get only the audio

  


  ffmpeg_cmd = "/opt/bin/ffmpeg -i \"" + s3_source_signed_url + "\" -f mp3 -ab 192000 -vn  -" ___ . 


  


  And it works as expected.

  


• How do I get event emitter based code to wait in nodejs ?

  29 mars 2019, par DigitalDisaster

  I may be asking the question wrong because I don’t have a lot of experience with this. Basically, I have this code :

  videoshow(images, videoOptions)
  
    .audio('song.mp3')
  
    .save('video.mp4')
  
    .on('start', function (command) {
  
      console.log('ffmpeg process started:', command)
  
    })
  
    .on('error', function (err, stdout, stderr) {
  
      console.error('Error:', err)
  
      console.error('ffmpeg stderr:', stderr)
  
    })
  
    .on('end', function (output) {
  
      console.error('Video created in:', output)
  
    })

  from this library

  It runs async as far as I understand, but I can’t get it to wait even when I add await before it.

  I want the block of code to stop until the end has finished and the image has been converted to a vide. I don’t see any examples, and I’m not sure what I’m doing wrong.