Recherche avancée

Sur d’autres sites (3369)

• Can FFmpeg crop gif image under 1 second ?

  5 septembre 2021, par jeongbalmountain

  I am developing an iOS application about gif.

  


  But I have a problem with gif cropping. Cropping gif spends so many times with FFmpeg.

  


  (updated code)

  


  These are the ffmpeg codes I've tried.

  


  cropped images -> gif

  


  ffmpeg -f image2 -i %d.jpg -vf 'scale=450:-1' output.gif


  


  using concat command (cropped images -> gif)

  


  ffmpeg -f concat -i tmp.txt output.gif


  


  using crop filter

  


  ffmpeg -i in.gif -filter:v "crop=out_w:out_h:x:y" out.gif


  


  These codes are so slow when converting images to gif
or gif to cropped gif.

  


  Is there a way to reduce the time to crop gif less than 1 second ?

  


  (updated2)
ffmpeg -i in.gif -filter:v "crop=out_w:out_h:x:y" out.gif

  


  


  ffmpeg version 4.4 Copyright (c) 2000-2021 the FFmpeg developers
  
built with Apple clang version 12.0.5 (clang-1205.0.22.9)
  
configuration : —prefix=/usr/local/Cellar/ffmpeg/4.4_2 —enable-shared

  


  


  


  —enable-pthreads —enable-version3 —cc=clang —host-cflags= —host-ldflags= —enable-ffplay —enable-gnutls —enable-gpl —enable-libaom —enable-libbluray —enable-libdav1d —enable-libmp3lame —enable-libopus —enable-librav1e —enable-librubberband —enable-libsnappy —enable-libsrt —enable-libtesseract —enable-libtheora —enable-libvidstab —enable-libvorbis —enable-libvpx —enable-libwebp —enable-libx264 —enable-libx265 —enable-libxml2 —enable-libxvid —enable-lzma —enable-libfontconfig —enable-libfreetype —enable-frei0r —enable-libass —enable-libopencore-amrnb —enable-libopencore-amrwb —enable-libopenjpeg —enable-libspeex —enable-libsoxr —enable-libzmq —enable-libzimg —disable-libjack —disable-indev=jack —enable-avresample —enable-videotoolbox libavutil 56. 70.100 / 56. 70.100 libavcodec 58.134.100 /
58.134.100 libavformat 58. 76.100 / 58. 76.100 libavdevice 58. 13.100 / 58. 13.100 libavfilter 7.110.100 / 7.110.100 libavresample 4. 0. 0 / 4. 0. 0 libswscale 5. 9.100 /
5. 9.100 libswresample 3. 9.100 / 3. 9.100 libpostproc 55. 9.100 / 55. 9.100 Input #0, gif, from './test.gif' : Duration : 00:00:18.27, start : 0.000000, bitrate : 70098 kb/s Stream #0:0 :
Video : gif, bgra, 1080x1920, 15 fps, 14.99 tbr, 100 tbn, 100 tbc
Stream mapping : Stream #0:0 -> #0:0 (gif (native) -> gif (native))
Press [q] to stop, [?] for help Output #0, gif, to './crop.gif' :
  
Metadata :
encoder : Lavf58.76.100 Stream #0:0 : Video : gif, bgr8(pc, progressive), 1080x1198, q=2-31, 200 kb/s, 14.99 fps, 100 tbn
Metadata :
encoder : Lavc58.134.100 gif frame= 1 fps=0.0 q=-0.0 size= 0kB time=00:00:00.01 bitrate= 0.0kbits/s speed=5e+03x

  


  


  


  frame= 11 fps=0.0 q=-0.0 size= 4096kB time=00:00:00.68
bitrate=49344.8kbits/s speed=1.31x frame= 22 fps= 21 q=-0.0
size= 8192kB time=00:00:01.41 bitrate=47594.9kbits/s speed=1.32x
  
frame= 32 fps= 20 q=-0.0 size= 12288kB time=00:00:02.08
bitrate=48395.8kbits/s speed=1.32x frame= 43 fps= 20 q=-0.0
size= 16384kB time=00:00:02.81 bitrate=47764.3kbits/s speed=1.33x
  
frame= 54 fps= 20 q=-0.0 size= 20992kB time=00:00:03.55
bitrate=48441.3kbits/s speed=1.33x frame= 65 fps= 20 q=-0.0
size= 24832kB time=00:00:04.28 bitrate=47528.9kbits/s speed=1.35x
  
frame= 77 fps= 21 q=-0.0 size= 28928kB time=00:00:05.08
bitrate=46649.2kbits/s speed=1.36x frame= 87 fps= 21 q=-0.0
size= 32768kB time=00:00:05.75 bitrate=46684.4kbits/s speed=1.36x
  
frame= 98 fps= 21 q=-0.0 size= 36608kB time=00:00:06.48
bitrate=46279.7kbits/s speed=1.36x frame= 109 fps= 21 q=-0.0
size= 40704kB time=00:00:07.22 bitrate=46183.8kbits/s speed=1.37x
  
frame= 120 fps= 21 q=-0.0 size= 44800kB time=00:00:07.95
bitrate=46163.7kbits/s speed=1.37x frame= 131 fps= 21 q=-0.0
size= 49152kB time=00:00:08.69 bitrate=46335.2kbits/s speed=1.38x
  
frame= 142 fps= 21 q=-0.0 size= 53504kB time=00:00:09.42
bitrate=46529.2kbits/s speed=1.37x frame= 153 fps= 21 q=-0.0
size= 57344kB time=00:00:10.15 bitrate=46282.0kbits/s speed=1.38x
  
frame= 164 fps= 21 q=-0.0 size= 61184kB time=00:00:10.89
bitrate=46025.6kbits/s speed=1.38x frame= 174 fps= 21 q=-0.0
size= 65280kB time=00:00:11.55 bitrate=46300.8kbits/s speed=1.37x
  
frame= 185 fps= 21 q=-0.0 size= 69376kB time=00:00:12.29
bitrate=46243.1kbits/s speed=1.38x frame= 196 fps= 21 q=-0.0
size= 72960kB time=00:00:13.02 bitrate=45905.4kbits/s speed=1.38x
  
frame= 207 fps= 21 q=-0.0 size= 76544kB time=00:00:13.76
bitrate=45570.4kbits/s speed=1.38x frame= 218 fps= 21 q=-0.0
size= 80640kB time=00:00:14.49 bitrate=45590.3kbits/s speed=1.38x
  
frame= 229 fps= 21 q=-0.0 size= 84224kB time=00:00:15.23
bitrate=45302.9kbits/s speed=1.39x frame= 240 fps= 21 q=-0.0
size= 88064kB time=00:00:15.96 bitrate=45201.8kbits/s speed=1.39x
  
frame= 250 fps= 21 q=-0.0 size= 91648kB time=00:00:16.63
bitrate=45146.1kbits/s speed=1.39x frame= 261 fps= 21 q=-0.0
size= 96000kB time=00:00:17.36 bitrate=45301.4kbits/s speed=1.38x
  
frame= 271 fps= 21 q=-0.0 size= 99840kB time=00:00:18.03
bitrate=45362.7kbits/s speed=1.38x frame= 274 fps= 21 q=-0.0
Lsize= 101552kB time=00:00:18.23 bitrate=45634.2kbits/s speed=1.37x
  
video:101552kB audio:0kB subtitle:0kB other streams:0kB global
headers:0kB muxing overhead : 0.000019%

  


  


  it takes 12 seconds

  


• Reasons for "Segmentation fault (core dumped)" when using Python extension and FFmpeg

  24 août 2021, par Christian Vorhemus

  I want to write a Python C extension that includes a function convertVideo() that converts a video from one format to another making use of FFmpeg 3.4.8 (the libav* libaries). The code of the extension is at the end of the question. The extension compiles successfully but whenever I open Python and want to call it (using a simple Python wrapper code that I don't include here), I get

  


  Python 3.7.10 (default, May  2 2021, 18:28:10)
[GCC 9.1.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import myModule
>>> myModule.convert_video("/home/admin/video.h264", "/home/admin/video.mp4")
convert 0
convert 1
Format raw H.264 video, duration -9223372036854775808 us
convert 2
Segmentation fault (core dumped)


  


  The interesting thing is, I wrote a simple helper program test_convert.cc that calls convertVideo() like so

  


  #include 
#include 

int convertVideo(const char *in_filename, const char *out_filename);

int main() {
  int res = convertVideo("/home/admin/video.h264", "/home/admin/video.mp4");
  return 0;
}


  


  and I compiled this program making use of the shared library that Python generates when building the C extension like so

  


  gcc test_convert.cc /usr/lib/python3.7/site-packages/_myModule.cpython-37m-aarch64-linux-gnu.so -o test_convert


  


  And it works ! The output is

  


  root# ./test_convert
convert 0
convert 1
Format raw H.264 video, duration -9223372036854775808 us
convert 2
convert 3
convert 4
convert 5
convert 6
Output #0, mp4, to '/home/admin/video.mp4':
    Stream #0:0: Video: h264 (High), yuv420p(tv, bt470bg, progressive), 1280x720 [SAR 1:1 DAR 16:9], q=2-31
convert 7


  


  The extension code looks like this

  


  #include 

#include 
#include 
#include 
#include 
#include 
#include 
#include 

extern "C"
{
#include "libavformat/avformat.h"
#include "libavutil/imgutils.h"
}

int convertVideo(const char *in_filename, const char *out_filename)
{
  // Input AVFormatContext and Output AVFormatContext
  AVFormatContext *input_format_context = avformat_alloc_context();
  AVPacket pkt;

  int ret, i;
  int frame_index = 0;
  printf("convert 0\n");
  av_register_all();
  printf("convert 1\n");
  // Input
  if ((ret = avformat_open_input(&input_format_context, in_filename, NULL,
                                 NULL)) < 0)
  {
    printf("Could not open input file.");
    return 1;
  }
  else
  {
    printf("Format %s, duration %lld us\n",
           input_format_context->iformat->long_name,
           input_format_context->duration);
  }
  printf("convert 2\n");
  if ((ret = avformat_find_stream_info(input_format_context, 0)) < 0)
  {
    printf("Failed to retrieve input stream information");
    return 1;
  }
  printf("convert 3\n");
  AVFormatContext *output_format_context = avformat_alloc_context();
  AVPacket packet;
  int stream_index = 0;
  int *streams_list = NULL;
  int number_of_streams = 0;
  int fragmented_mp4_options = 0;
  printf("convert 4\n");
  avformat_alloc_output_context2(&output_format_context, NULL, NULL,
                                 out_filename);
  if (!output_format_context)
  {
    fprintf(stderr, "Could not create output context\n");
    ret = AVERROR_UNKNOWN;
    return 1;
  }
  printf("convert 5\n");
  AVOutputFormat *fmt = av_guess_format(0, out_filename, 0);
  output_format_context->oformat = fmt;

  number_of_streams = input_format_context->nb_streams;
  streams_list =
      (int *)av_mallocz_array(number_of_streams, sizeof(*streams_list));

  if (!streams_list)
  {
    ret = AVERROR(ENOMEM);
    return 1;
  }
  for (i = 0; i < input_format_context->nb_streams; i++)
  {
    AVStream *out_stream;
    AVStream *in_stream = input_format_context->streams[i];
    AVCodecParameters *in_codecpar = in_stream->codecpar;
    if (in_codecpar->codec_type != AVMEDIA_TYPE_AUDIO &&
        in_codecpar->codec_type != AVMEDIA_TYPE_VIDEO &&
        in_codecpar->codec_type != AVMEDIA_TYPE_SUBTITLE)
    {
      streams_list[i] = -1;
      continue;
    }
    streams_list[i] = stream_index++;

    out_stream = avformat_new_stream(output_format_context, NULL);
    if (!out_stream)
    {
      fprintf(stderr, "Failed allocating output stream\n");
      ret = AVERROR_UNKNOWN;
      return 1;
    }
    ret = avcodec_parameters_copy(out_stream->codecpar, in_codecpar);
    if (ret < 0)
    {
      fprintf(stderr, "Failed to copy codec parameters\n");
      return 1;
    }
  }
  printf("convert 6\n");
  av_dump_format(output_format_context, 0, out_filename, 1);
  if (!(output_format_context->oformat->flags & AVFMT_NOFILE))
  {
    ret = avio_open(&output_format_context->pb, out_filename, AVIO_FLAG_WRITE);
    if (ret < 0)
    {
      fprintf(stderr, "Could not open output file '%s'", out_filename);
      return 1;
    }
  }
  AVDictionary *opts = NULL;
  printf("convert 7\n");
  ret = avformat_write_header(output_format_context, &opts);
  if (ret < 0)
  {
    fprintf(stderr, "Error occurred when opening output file\n");
    return 1;
  }
  int n = 0;

  while (1)
  {
    AVStream *in_stream, *out_stream;
    ret = av_read_frame(input_format_context, &packet);
    if (ret < 0)
      break;
    in_stream = input_format_context->streams[packet.stream_index];
    if (packet.stream_index >= number_of_streams ||
        streams_list[packet.stream_index] < 0)
    {
      av_packet_unref(&packet);
      continue;
    }
    packet.stream_index = streams_list[packet.stream_index];

    out_stream = output_format_context->streams[packet.stream_index];

    out_stream->codec->time_base.num = 1;
    out_stream->codec->time_base.den = 30;

    packet.pts = n * 3000;
    packet.dts = n * 3000;
    packet.duration = 3000;

    packet.pos = -1;

    ret = av_interleaved_write_frame(output_format_context, &packet);
    if (ret < 0)
    {
      fprintf(stderr, "Error muxing packet\n");
      break;
    }
    av_packet_unref(&packet);
    n++;
  }

  av_write_trailer(output_format_context);
  avformat_close_input(&input_format_context);
  if (output_format_context &&
      !(output_format_context->oformat->flags & AVFMT_NOFILE))
    avio_closep(&output_format_context->pb);
  avformat_free_context(output_format_context);
  av_freep(&streams_list);
  if (ret < 0 && ret != AVERROR_EOF)
  {
    fprintf(stderr, "Error occurred\n");
    return 1;
  }
  return 0;
}
// PyMethodDef and other orchestration code is skipped


  


  What is the reason that the code works as expected in my test_convert but not within Python ?

  


• FFMPEG encoding 16bit video data results in 10bit

  12 mars 2023, par Jl arto

  I want to compress a depth map that has 16 bits of information per pixel. In general, such depth maps can be stored in different ways, e.g. p016le, gray16le, yuv420p16le, yuv444p16le, ... but for simplicity, let's assume the depth map is a yuv420p16le (where the y-channel contains the depth).

  


  For some reason when encoding with hevc_nvenc (I use an NVIDIA GTX 1660 Ti GPU), ffmpeg (the command line tool) changes the pixel format to a 10 or 12 bit variant (p010le, gray12le, yuv420p10le, yuv444p12le, ...), but I would like to keep the full 16 bits, since this affects the quality of the depth stored.

  


  For example :

  


  ffmpeg.exe -s:v 1920x1080 -r 30 -pix_fmt yuv420p16le -i depth_yuv420p16le.yuv -c:v hevc_nvenc -pix_fmt yuv444p16le output.mp4


  


  If I use ffprobe on the output.mp4, it tells me that the underlying pixel format is actually yuv444p10le. (Decoding and looking at the raw pixel data, I can confirm that the precision has decreased from 16 bits to 10 bits).

  


  I hope 16 bit compression is possible, since according to

  


  ffmpeg -h encoder=hevc_nvenc


  


  the supported pixel formats are :

  


  hevc_nvenc: yuv420p nv12 p010le yuv444p p016le yuv444p16le bgr0 rgb0 cuda d3d11


  


  But p016le results in a p010le output, and yuv444p16le in yuv444p10le.

  


  Does anyone know where the problem could lie ? Should I re-install ffmpeg (version 4.3.2-2021-02-27-essentials_build-www.gyan.dev) ? Is it because of Windows 10 having limited encoding/decoding capabilities ? Will buying the HEVC Video Extensions help solve this problem ?

  


  Additional info : using libx256 does not look like it will work for this purpose, since the supported pixel formats are :

  


  libx256 : yuv420p yuvj420p yuv422p yuvj422p yuv444p yuvj444p gbrp yuv420p10le yuv422p10le yuv444p10le gbrp10le yuv420p12le yuv422p12le yuv444p12le gbrp12le gray gray10le gray12le


  


  Any help would be greatly appreciated.