Recherche avancée

Sur d’autres sites (12141)

• Transcoding HEVC 4K HDR using ffmpeg and vaapi

  13 mai 2021, par P. Brand

  I am trying to transcode a 4K HEVC HDR video to a lower bitrate using ffmpeg and VAAPI hardware acceleration for my LG OLED television.

  



  I use the following file :
Video

  



  Using the following command :

  



  ffmpeg -init_hw_device vaapi=va:/dev/dri/renderD128 -hwaccel vaapi -hwaccel_output_format vaapi -hwaccel_device va -filter_hw_device va -i LG\ 4K\ HDR\ Demo\ -\ Daylight.mkv -map 0:0 -t 00:00:20 -c:v hevc_vaapi -sei hdr -qp:v 21 lg_vaapi.mkv


  



  But this file is not recognized as HDR on my TV.

  



  When using the libx265 encoder (which is off course much slower) it works fine :

  



  ffmpeg -i LG\ 4K\ HDR\ Demo\ -\ Daylight.mkv -t 00:00:20 -map 0:0 -c:v libx265 -preset medium -crf 18 -x265-params "colorprim=bt2020:colormatrix=bt2020nc:transfer=smpte2084:colormatrix=bt2020nc:hdr=1:info=1:repeat-headers=1:master-display=G(13250,34500)B(7500,3000)R(34000,16000)WP(15635,16450)L(11000000,40)" lg_libx265.mkv


  



  When comparing the output of mediainfo on both files I can see the following properties are not set for the video track :

  



  

  • Color range

  


  • Color primaries

  


  • Transfer characteristics

  


  



  mediainfo for vaapi encode :

  



  General
Unique ID                                : 168011494166392912924249315217763643529 
(0x7E65D636029A5354FE73998A5ED6B089)
Complete name                            : lg_vaapi.mkv
Format                                   : Matroska
Format version                           : Version 4
File size                                : 118 MiB
Duration                                 : 20 s 4 ms
Overall bit rate                         : 49.6 Mb/s
Writing application                      : Lavf58.20.100
Writing library                          : Lavf58.20.100
ErrorDetectionType                       : Per level 1

Video
ID                                       : 1
Format                                   : HEVC
Format/Info                              : High Efficiency Video Coding
Format profile                           : Main 10@L5@Main
HDR format                               : SMPTE ST 2086, HDR10 compatible
Codec ID                                 : V_MPEGH/ISO/HEVC
Duration                                 : 20 s 4 ms
Bit rate                                 : 48.6 Mb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate mode                          : Constant
Frame rate                               : 59.940 (60000/1001) FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0
Bit depth                                : 10 bits
Bits/(Pixel*Frame)                       : 0.098
Stream size                              : 116 MiB (98%)
Writing library                          : Lavc58.35.100 hevc_vaapi
Default                                  : Yes
Forced                                   : No
Mastering display color primaries        : Display P3
Mastering display luminance              : min: 0.0040 cd/m2, max: 1100 cd/m2


  



  mediainfo for the libx265 encode :

  



  General
Unique ID                                : 261605749278874425324960669956061374077 (0xC4CF6F42FE952004772A02D5819DA67D)
Complete name                            : lg_libx265.mkv
Format                                   : Matroska
Format version                           : Version 4
File size                                : 155 MiB
Duration                                 : 47 s 715 ms
Overall bit rate                         : 27.3 Mb/s
Writing application                      : Lavf58.20.100
Writing library                          : Lavf58.20.100
ErrorDetectionType                       : Per level 1

Video
ID                                       : 1
Format                                   : HEVC
Format/Info                              : High Efficiency Video Coding
Format profile                           : Main 10@L5.1@Main
HDR format                               : SMPTE ST 2086, HDR10 compatible
Codec ID                                 : V_MPEGH/ISO/HEVC
Duration                                 : 47 s 715 ms
Bit rate                                 : 26.8 Mb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate mode                          : Constant
Frame rate                               : 59.940 (60000/1001) FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0
Bit depth                                : 10 bits
Bits/(Pixel*Frame)                       : 0.054
Stream size                              : 152 MiB (98%)
Writing library                          : x265 2.8:[Linux][GCC 9.1.0][64 bit] 10bit
Encoding settings                        : cpuid=1111039 / frame-threads=3 / wpp / no-pmode / no-pme / no-psnr / no-ssim / log-level=2 / input-csp=1 / input-res=3840x2160 / interlace=0 / total-frames=0 / level-idc=0 / high-tier=1 / uhd-bd=0 / ref=3 / no-allow-non-conformance / repeat-headers / annexb / no-aud / no-hrd / info / hash=0 / no-temporal-layers / open-gop / min-keyint=25 / keyint=250 / gop-lookahead=0 / bframes=4 / b-adapt=2 / b-pyramid / bframe-bias=0 / rc-lookahead=20 / lookahead-slices=8 / scenecut=40 / radl=0 / no-intra-refresh / ctu=64 / min-cu-size=8 / no-rect / no-amp / max-tu-size=32 / tu-inter-depth=1 / tu-intra-depth=1 / limit-tu=0 / rdoq-level=0 / dynamic-rd=0.00 / no-ssim-rd / signhide / no-tskip / nr-intra=0 / nr-inter=0 / no-constrained-intra / strong-intra-smoothing / max-merge=2 / limit-refs=3 / no-limit-modes / me=1 / subme=2 / merange=57 / temporal-mvp / weightp / no-weightb / no-analyze-src-pics / deblock=0:0 / sao / no-sao-non-deblock / rd=3 / no-early-skip / rskip / no-fast-intra / no-tskip-fast / no-cu-lossless / no-b-intra / no-splitrd-skip / rdpenalty=0 / psy-rd=2.00 / psy-rdoq=0.00 / no-rd-refine / no-lossless / cbqpoffs=0 / crqpoffs=0 / rc=crf / crf=18.0 / qcomp=0.60 / qpstep=4 / stats-write=0 / stats-read=0 / ipratio=1.40 / pbratio=1.30 / aq-mode=1 / aq-strength=1.00 / cutree / zone-count=0 / no-strict-cbr / qg-size=32 / no-rc-grain / qpmax=69 / qpmin=0 / no-const-vbv / sar=1 / overscan=0 / videoformat=5 / range=0 / colorprim=9 / transfer=16 / colormatrix=9 / chromaloc=0 / display-window=0 / master-display=G(13250,34500)B(7500,3000)R(34000,16000)WP(15635,16450)L(10000000,50) / max-cll=0,0 / min-luma=0 / max-luma=1023 / log2-max-poc-lsb=8 / vui-timing-info / vui-hrd-info / slices=1 / no-opt-qp-pps / no-opt-ref-list-length-pps / no-multi-pass-opt-rps / scenecut-bias=0.05 / no-opt-cu-delta-qp / no-aq-motion / hdr / no-hdr-opt / no-dhdr10-opt / no-idr-recovery-sei / analysis-reuse-level=5 / scale-factor=0 / refine-intra=0 / refine-inter=0 / refine-mv=0 / no-limit-sao / ctu-info=0 / no-lowpass-dct / refine-mv-type=0 / copy-pic=1 / max-ausize-factor=1.0 / no-dynamic-refine / no-single-sei
Default                                  : Yes
Forced                                   : No
Color range                              : Limited
Color primaries                          : BT.2020
Transfer characteristics                 : PQ
Matrix coefficients                      : BT.2020 non-constant
Mastering display color primaries        : Display P3
Mastering display luminance              : min: 0.0040 cd/m2, max: 1100 cd/m2


  



  I suspect Color range, Color primaries and Transfer characteristics are necessary to make the file recognized as HDR by my TV. 
How can I add these properties to the VAAPI encoded file ?

  



  Edit :
This patch fixes the problem :
Patch

  


• How to extract elementary video from mp4 using ffmpeg programmatically ?

  24 octobre 2019, par epipav

  I have started learning ffmpeg few weaks ago. At the moment I am able to transcode any video to mp4 using h264/AVC codec. The main scheme is something like that :

  -open input
  
  -demux
  
  -decode
  
  -encode
  
  -mux

  The actual code is below :

  #include <iostream>
  
  
  
  #include 
  
  
  
  extern "C" {
  
  
  
    #
  
    ifndef __STDC_CONSTANT_MACROS# undef main /* Prevents SDL from overriding main() */ # define __STDC_CONSTANT_MACROS# endif
  
  
  
    # pragma comment(lib, "avcodec.lib")# pragma comment(lib, "avformat.lib")# pragma comment(lib, "swscale.lib")# pragma comment(lib, "avutil.lib")
  
  
  
    #include 
  
    #include 
  
    #include 
  
    #include 
  
    #include <libavutil></libavutil>opt.h>
  
    #include 
  
    #include 
  
    #include 
  
    #include 
  
    #include 
  
  }
  
  
  
  using namespace std;
  
  
  
  void open_video(AVFormatContext * oc, AVCodec * codec, AVStream * st) {
  
    int ret;
  
    AVCodecContext * c;
  
    c = st - > codec;
  
  
  
    /*open codec */
  
  
  
    cout &lt;&lt; "probably starts here" &lt;&lt; endl;
  
    ret = avcodec_open2(c, codec, NULL);
  
    cout &lt;&lt; "and ends here" &lt;&lt; endl;
  
  
  
    if (ret &lt; 0) {
  
      cout &lt;&lt; ("Could not open video codec") &lt;&lt; endl;
  
    }
  
  
  
  }
  
  
  
  /*This function will add a new stream to our file.
  
  @param
  
  oc -> Format context that the new stream will be added.
  
  codec -> codec of the stream, this will be passed.
  
  codec_id -> 
  
  chWidth->
  
  chHeight->
  
  */
  
  
  
  AVStream * addStream(AVFormatContext * oc, AVCodec ** codec, enum AVCodecID codec_id, int chWidth, int chHeight, int fps) {
  
    AVCodecContext * c;
  
    AVStream * st;
  
  
  
    //find encoder of the stream, it passes this information to @codec, later on
  
    //it will be used in encoding the video @ avcodec_encode_video2 in loop.
  
    * codec = avcodec_find_encoder(AV_CODEC_ID_H264);
  
  
  
    if (( * codec) == NULL)
  
      cout &lt;&lt; "ERROR CAN NOT FIND ENCODER! ERROR! ERROR! AVCODEC_FIND_ENCODER FAILED !!!1 "
  
    "" &lt;&lt; endl;
  
  
  
    if (!( * codec))
  
      printf("Could not find encoder for ' %s ' ", avcodec_get_name(codec_id));
  
  
  
    //create a new stream with the found codec inside oc(AVFormatContext).
  
    st = avformat_new_stream(oc, * codec);
  
  
  
    if (!st)
  
      cout &lt;&lt; " Cannot allocate stream " &lt;&lt; endl;
  
  
  
    //Setting the stream id.
  
    //Since, there can be other streams in this AVFormatContext,
  
    //we should find the first non used index. And this is oc->nb_streams(number of streams) - 1
  
    st - > id = oc - > nb_streams - 1;
  
  
  
    c = st - > codec;
  
  
  
    //setting the stream's codec's properties.
  
    c - > codec_id = codec_id;
  
    c - > bit_rate = 4000000;
  
    c - > width = chWidth;
  
    c - > height = chHeight;
  
    c - > time_base.den = fps;
  
    //fps;
  
    c - > time_base.num = 1;
  
    c - > gop_size = 12;
  
    c - > pix_fmt = AV_PIX_FMT_YUV420P;
  
  
  
    if (c - > codec_id == AV_CODEC_ID_MPEG2VIDEO) {
  
      /* just for testing, we also add B frames */
  
      c - > max_b_frames = 2;
  
    }
  
  
  
    if (c - > codec_id == AV_CODEC_ID_MPEG1VIDEO) {
  
      /* Needed to avoid using macroblocks in which some coeffs overflow.
  
       * This does not happen with normal video, it just happens here as
  
       * the motion of the chroma plane does not match the luma plane. */
  
      c - > mb_decision = 2;
  
    }
  
  
  
    /* Some formats want stream headers to be separate. */
  
    if (oc - > oformat - > flags &amp; AVFMT_GLOBALHEADER)
  
      c - > flags |= CODEC_FLAG_GLOBAL_HEADER;
  
  
  
    //returning our lovely new brand stream.
  
    return st;
  
  
  
  }
  
  
  
  int changeResolution(string source, int format) {
  
    //Data members
  
    struct SwsContext * sws_ctx = NULL;
  
    AVFrame * pFrame = NULL;
  
    AVFrame * outFrame = NULL;
  
    AVPacket packet;
  
    uint8_t * buffer = NULL;
  
    uint8_t endcode[] = {
  
      0,
  
      0,
  
      1,
  
      0xb7
  
    };
  
    AVDictionary * optionsDict = NULL;
  
    AVFormatContext * pFormatCtx = NULL;
  
    AVFormatContext * outputContext = NULL;
  
    AVCodecContext * pCodecCtx;
  
    AVCodec * pCodec;
  
    AVCodec * codec;
  
    AVCodec * videoCodec;
  
    AVOutputFormat * fmt;
  
    AVStream * video_stream;
  
    int changeWidth;
  
    int changeHeight;
  
    int frameFinished;
  
    int numBytes;
  
    int fps;
  
  
  
    int lock = 0;
  
  
  
    //Register all codecs &amp; other important stuff. Vital!..
  
    av_register_all();
  
  
  
    //Selects the desired resolution.
  
    if (format == 0) {
  
      changeWidth = 320;
  
      changeHeight = 180;
  
    } else if (format == 1) {
  
      changeWidth = 640;
  
      changeHeight = 480;
  
  
  
    } else if (format == 2) {
  
      changeWidth = 960;
  
      changeHeight = 540;
  
  
  
    } else if (format == 3) {
  
      changeWidth = 1024;
  
      changeHeight = 768;
  
  
  
    } else {
  
      changeWidth = 1280;
  
      changeHeight = 720;
  
    }
  
  
  
    // Open video file
  
    int aaa;
  
    aaa = avformat_open_input( &amp; pFormatCtx, source.c_str(), NULL, NULL);
  
    if (aaa != 0) {
  
      cout &lt;&lt; " cannot open input file \n" &lt;&lt; endl;
  
      cout &lt;&lt; "aaa = " &lt;&lt; aaa &lt;&lt; endl;
  
      return -1; // Couldn't open file    
  
    }
  
  
  
    // Retrieve stream information
  
    if (av_find_stream_info(pFormatCtx) &lt; 0)
  
      return -1; // Couldn't find stream information
  
  
  
    //just checking duration casually for no reason
  
    /*int64_t duration = pFormatCtx->duration;
  
  
  
    cout &lt;&lt; "the duration is " &lt;&lt; duration &lt;&lt; " " &lt;&lt; endl;*/
  
  
  
    //this writes the info about the file
  
    av_dump_format(pFormatCtx, 0, 0, 0);
  
    cin >> lock;
  
  
  
    // Find the first video stream
  
    int videoStream = -1;
  
    int i;
  
  
  
    for (i = 0; i &lt; 3; i++)
  
      if (pFormatCtx - > streams[i] - > codec - > codec_type == AVMEDIA_TYPE_VIDEO) {
  
        videoStream = i;
  
        cout &lt;&lt; " lel \n ";
  
        break;
  
  
  
      }
  
  
  
    if (videoStream == -1)
  
      return -1; // Didn't find a video stream
  
  
  
    // Get a pointer to the codec context for the video stream
  
    pCodecCtx = pFormatCtx - > streams[videoStream] - > codec;
  
    fps = pCodecCtx - > time_base.den;
  
  
  
    //Find the decoder of the input file, for the video stream
  
    pCodec = avcodec_find_decoder(pCodecCtx - > codec_id);
  
  
  
    if (pCodec == NULL) {
  
      fprintf(stderr, "Unsupported codec!\n");
  
      return -1; // Codec not found
  
    }
  
  
  
    // Open codec, you must open it first, in order to use it.
  
    if (avcodec_open2(pCodecCtx, pCodec, &amp; optionsDict) &lt; 0)
  
      return -1; // Could not open codec
  
  
  
    // Allocate video frame ( pFrame for taking the packets into, outFrame for processed frames to packet.)
  
    pFrame = avcodec_alloc_frame();
  
    outFrame = avcodec_alloc_frame();
  
  
  
    i = 0;
  
  
  
    int ret;
  
    int video_frame_count = 0;
  
  
  
    //Initiate the outFrame set the buffer &amp; fill the properties
  
    numBytes = avpicture_get_size(PIX_FMT_YUV420P, changeWidth, changeHeight);
  
    buffer = (uint8_t * ) av_malloc(numBytes * sizeof(uint8_t));
  
    avpicture_fill((AVPicture * ) outFrame, buffer, PIX_FMT_YUV420P, changeWidth, changeHeight);
  
  
  
    int pp;
  
    int frameNo = 0;
  
  
  
    //allocate the outputContext, it will be the AVFormatContext of our output file. 
  
    //It will try to find the format by giving the file name.
  
    avformat_alloc_output_context2( &amp; outputContext, NULL, NULL, "myoutput.mp4");
  
  
  
    //Cant find the file extension, using MPEG as default.
  
    if (!outputContext) {
  
      printf("Could not deduce output format from file extension: using MPEG.\n");
  
      avformat_alloc_output_context2( &amp; outputContext, NULL, "mpeg", "myoutput.mp4");
  
    }
  
  
  
    //Still cant set file extension, exit.
  
    if (!outputContext) {
  
      return 1;
  
    }
  
  
  
    //set AVOutputFormat fmt to our outputContext's format.
  
    fmt = outputContext - > oformat;
  
    video_stream = NULL;
  
  
  
    //If fmt has a valid codec_id, create a new video stream.
  
    //This function will set the streams codec &amp; codecs desired properties.
  
    //Stream's codec will be passed to videoCodec for later usage.
  
    if (fmt - > video_codec != AV_CODEC_ID_NONE)
  
      video_stream = addStream(outputContext, &amp; videoCodec, fmt - > video_codec, changeWidth, changeHeight, fps);
  
  
  
    //open the video using videoCodec. by avcodec_open2() i.e open the codec.
  
    if (video_stream)
  
      open_video(outputContext, videoCodec, video_stream);
  
  
  
    //Creating our new output file.
  
    if (!(fmt - > flags &amp; AVFMT_NOFILE)) {
  
      ret = avio_open( &amp; outputContext - > pb, "toBeStreamed.264", AVIO_FLAG_WRITE);
  
      if (ret &lt; 0) {
  
        cout &lt;&lt; " cant open file " &lt;&lt; endl;
  
        return 1;
  
      }
  
    }
  
  
  
    //Writing the header of format context.
  
    //ret = avformat_write_header(outputContext, NULL);
  
  
  
    if (ret >= 0) {
  
      cout &lt;&lt; "writing header success !!!" &lt;&lt; endl;
  
    }
  
  
  
    //Start reading packages from input file.
  
    while (av_read_frame(pFormatCtx, &amp; packet) >= 0) {
  
  
  
      // Is this a packet from the video stream?  
  
      if (packet.stream_index == videoStream) {
  
  
  
        // Decode video package into frames
  
        ret = avcodec_decode_video2(pCodecCtx, pFrame, &amp; frameFinished, &amp; packet);
  
  
  
        if (ret &lt; 0) {
  
          printf(" Error decoding frame !!..");
  
          return ret;
  
        }
  
  
  
        if (frameFinished) {
  
          printf("video_frame n:%d    coded_n:%d\n", video_frame_count++, pFrame - > coded_picture_number);
  
        }
  
  
  
        av_free_packet( &amp; packet);
  
  
  
        //do stuff with frame, in this case we are changing the resolution.
  
        static struct SwsContext * img_convert_ctx_in = NULL;
  
        if (img_convert_ctx_in == NULL) {
  
          img_convert_ctx_in = sws_getContext(pCodecCtx - > width,
  
            pCodecCtx - > height,
  
            pCodecCtx - > pix_fmt,
  
            changeWidth,
  
            changeHeight,
  
            PIX_FMT_YUV420P,
  
            SWS_BICUBIC,
  
            NULL,
  
            NULL,
  
            NULL);
  
  
  
        }
  
        //scale the frames
  
        sws_scale(img_convert_ctx_in,
  
          pFrame - > data,
  
          pFrame - > linesize,
  
          0,
  
          pCodecCtx - > height,
  
          outFrame - > data,
  
          outFrame - > linesize);
  
  
  
        //initiate the pts value
  
        if (frameNo == 0)
  
          outFrame - > pts = 0;
  
  
  
        //calculate the pts value &amp; set it.
  
        outFrame - > pts += av_rescale_q(1, video_stream - > codec - > time_base, video_stream - > time_base);
  
  
  
        //encode frames into packages. Package passed in @packet.
  
        if (avcodec_encode_video2(outputContext - > streams[0] - > codec, &amp; packet, outFrame, &amp; pp) &lt; 0)
  
          cout &lt;&lt; "Encoding frames into packages, failed. " &lt;&lt; endl;
  
  
  
        frameNo++;
  
  
  
        //write the packages into file, resulting in creating a video file.
  
        av_interleaved_write_frame(outputContext, &amp; packet);
  
  
  
      }
  
  
  
    }
  
  
  
    av_free_packet( &amp; packet);
  
    //av_write_trailer(outputContext);
  
  
  
    avio_close(outputContext - > pb);
  
  
  
    // Free the RGB image
  
    av_free(buffer);
  
    av_free(outFrame);
  
  
  
    // Free the YUV frame
  
    av_free(pFrame);
  
  
  
    // Close the codec
  
    avcodec_close(video_stream - > codec);
  
    avcodec_close(pCodecCtx);
  
  
  
    // Close the video file
  
    avformat_close_input( &amp; pFormatCtx);
  
  
  
    return 0;
  
  }
  
  </iostream>

  at the end of the process I get my desired file with desired codec & container & resolution.

  My problem is, in a part of our project I need to get elementary video streams IN file. Such as example.264. However I can not add a stream without creating an AVFormatContext. I can not create an AVFormatContext because 264 files does not have a container,they are just raw video ?, as far as I know.

  I have tried the way in decoding_encoding.c which uses fwrite. However that example was for mpeg-2 codec and when I try to adapt that code to H264/AVC codec, I got "floating point division by zero" error from mediainfo and moreover, some of the properties of the video was not showing (such as FPS & playtime & quality factor). I think it has to do with the "endcode" the example adds at the end of the code. It is for mpeg-2. ( uint8_t endcode[] = 0, 0, 1, 0xb7  ; )

  Anyway, I would love to get a startpoint for this task. I have managed to come this far by using internet resources ( quite few & outdated for ffmpeg) but now I’m stuck a little.

• FFMPEG : Set output res and bitrate depending upon input video

  21 octobre 2019, par Jegschemesch

  I’m processing user videos with differing aspect ratios. It seems FFMPEG only allows you to specify a fixed resolution. I want the output res to be appropriate for the input res. Similarly, I’d like FFMPEG to intelligently set the output bitrate based on the input video : obviously it shouldn’t be any bigger than the input.

  I can get the properties of a video with,

  ffmpeg -i example.flv

  But this requires some ugly parsing of the output, so I’m wondering if FFMPEG or some other tool has a more direct facility.

  Basically, I have the Youtube problem : crap comes in, reasonably uniform quality should come out.